Bad as in terrible, not bad as in really good

January 19, 2011 12:48 PM

I have an opportunity to give the most terrible scientific presentation I can possibly muster. Tell me how to make it awesomely bad.

Slide design I feel pretty confident about - unrelated clip art, disappearing information, tiny font, blocks of text, and so on. In terms of talk organization, data presentation, and delivery, I would like to fill my talk with every terrible thing that you hate about scientific presentations.

Some examples of things I intend to do:

* Have far to many slides and rush through the last 10 or so in two minutes
* Turn my back on the class to read slides
* No "signposting" slides
* Bad graphs

To hopefully keep this from degenerating into rant-swapping about terrible talks, please keep in mind I'm looking for things that I can successfully implement in a 20 minute talk, and which are instructive about what a good presentation should be like. Mild shenanigans are okay (I'm planning to spend the first minute in front of the class trying to open my presentation file) but let's keep it focused on things that inhibit the transfer of knowledge and make talks incomprehensible.

posted by heyforfour to science & nature (145 answers total)

Lots of jargon with little or, preferably, no explanation. Double points if you can use abbreviations this way too.
posted by theichibun at 12:50 PM on January 19, 2011 [6 favorites]

trail... off... at... every... opportunity....
posted by Neekee at 12:51 PM on January 19, 2011 [1 favorite]

Lots of equations. The more complicated the better.

Also make sure you say 'I won't read this slide' then go ahead and read it.

Have ill-fitting clothes on.
posted by chiefthe at 12:51 PM on January 19, 2011 [10 favorites]

Have your talk written down on A4 paper. Hold the paper in front of you and read it in a hushed monotone. Lose your place in the slides, so you have to backtrack.
posted by Paragon at 12:53 PM on January 19, 2011

Use lots of text when a simple graphic would do.

Read the slides as they are with no paraphrasing or elaboration.

Have tons of graphs or other data all on one slide so that they are too small for the audience to read.

Go through every graph or piece of data point by point in exhausting detail.

Don't summarize or give a "take home message" after discussing data. Let the audience figure out the point of your experiment for themselves.

Don't give any rationale for why you are doing the experiments. Especially no background information on why your research is important.

Jump right in to the data without giving background information.

Acronyms, acronyms, acronyms! You can never have enough. Don't define them, just assume everyone knows what they stand for.
posted by Knowyournuts at 12:55 PM on January 19, 2011 [1 favorite]

Have a long (five minutes or longer) video that you treat as self-explanatory. Stop in the middle of the talk so everyone can watch it. Works even better if you start it up right at the end of your allotted time.

True story.
posted by Pickman's Next Top Model at 12:55 PM on January 19, 2011 [3 favorites]

One unforgettable detail of the worst presentation I ever saw was a single slide with several graphs and charts of wildly differing scale. You might want to take a look at How To Lie With Statistics for some good examples of worst practices for data.

Also, at some point you should accidentally go out of slideshow mode and struggle with the powerpoint interface to get it back (ideally waiting for someone from the crowd to yell instructions).
posted by activitystory at 12:56 PM on January 19, 2011

Emulate, as much as possible, the presentation of the Rockwell Turbo Encabulator.
posted by namewithoutwords at 12:58 PM on January 19, 2011 [8 favorites]

After you finally find out how to open the presentation, go through the first few slides in "preparation" for yourself. Then, forget to set it back to the first slide when you're all ready.

Start at whichever slide you've left up.
posted by MustardTent at 12:59 PM on January 19, 2011

* Have some sort of "interesting" real-life demonstration or exhibit, but make sure it's way too small for anyone to see what's going on.

* Have a picture that "doesn't work" on the projector's computer, so you have to draw it on a chalkboard or on a piece of paper. Draw it really, really badly - the instructive part is that we should test our presentations on the auditorium computer before the big day.
posted by muddgirl at 12:59 PM on January 19, 2011

Include a few slides that show the covers of the prestigious journals you published in. Do not explain this work at all - assume that everyone in the audience read and loved it.
posted by Jorus at 1:00 PM on January 19, 2011

Have your slides out of order.

Switch randomly between your computer slides and some overhead transparencies pointed at the same projection screen. Make sure the transparencies are bad xeroxes of something that has already been mimeo'd to within an inch of its life.

For bonus points, when you need to make a clarifying comment, pull up the projection screen temporarily and scrawl on the chalkboard illegibly without turning off the projector or the overhead, then drop the screen back over your clarification before anyone can read it.

No I'm not still bitter about a guy who actually does this, why do you ask?
posted by dorque at 1:00 PM on January 19, 2011 [2 favorites]

-- About half-way through, with no warning or explanation, change your shirt as you are talking. Or shoes if you don't want to get all chesty about it. If the former, have someone write "She Blinded Me With Science" on your back, as close to a cool tattoo looking as possible, and turn around and strike a pose before you put the new shirt on.

-- if you can wrangle an accomplice, have some sort of food delivered in the middle, pay him etc, and never break stride, or crack a grin.
posted by timsteil at 1:01 PM on January 19, 2011 [2 favorites]

A slightly-more-subtle mistake that many students and postdocs make (especially in interview talks) is to present all of your work chronologically. Forget the concept of a clear narrative.

Also, use animations to transition between EVERYTHING.
posted by JMOZ at 1:02 PM on January 19, 2011

* Lots and lots of equations, none of them explained (or very relevant)
* Spending the first five minutes talking about yourself, your kids, etc (seriously, I have seen this a few times)
* The whole talk should be about some method or idea that you have made a cute abbreviation out of (e.g., SMART) but never once explain - not even to clarify that it IS a method or idea
* Make statements that are obviously incorrect logical fallacies (e.g., "It is getting colder in Kansas therefore global warming is not happening")
* Spend most of the time mumbling, but occasionally have bursts of loudness!and!excitement! for no apparent reason
* Have a slide that looks like a conclusion slide about 2/3 of the way through, thus raising everyone's hopes, only to dash them by going on with further incomprehensible mumbling
* Never stating the point of the talk or putting it into a large context, just diving right into method (or, worse, results)
* Have one slide with lots and lots of tiny graphs, with no apparent reason how they differ or what they show
posted by forza at 1:02 PM on January 19, 2011

Slide design things:

- Choose a really lame background with a color that does not show up well when projected
- Use Comic Sans font for body text
- Multi colored text
- Supposedly funny cartoon that has no relevance to the talk (supposedly, again, to "break the ice")
- Use words like basically, for example, invented etc multiple times on the same slide
- Spelling and grammar mistakes - these are not the one-off issues that all of us have, but really terrible ones. During the presentation, discover these mistakes and apologize


Talk:

- Don't have an "objectives" or "take away" section. Just jump in to the presentation
- After a couple of slides, remember to introduce yourself and go around the room (if there are less that 10 people)
- On a slide, jump ahead or refer an earlier slide. Do this as often as you can. When you do this, say "As we saw on slide X and slide Y" or "more details in the coming slides", as the case may be
- You have already covered reading the slides, which is the best bet!
- "Get back to them" for almost every question
- Instead of just rushing through the last 10 slides, vary it. After about 15 slides, look at your watch, mentally do the math and see that you are behind. Then you accelerate and after flipping through key slides, you realize that you now have ample time and can spend 10 minutes on one slide
- Take questions and when answering them, go off a tangent. When you come back, say "Where were we?"

Guaranteed failure in any presentation for choosing 2 or more of each section above!
posted by theobserver at 1:02 PM on January 19, 2011 [3 favorites]

Embedded media clips which do not play, and you have to futz around with the slide before proceeding, saying something like "If this had worked you would have seen X."

Introduce someone as the panel moderator, who will time you by holding up pieces of paper with 5 - 2 - 1 - 0 minutes on them as a countdown. When s/he shows the '0' sign, say "just one more minute please" in a slightly aggressive voice and then go on for another five minutes.
posted by carter at 1:03 PM on January 19, 2011 [1 favorite]

Include a slide that says "Introduce myself."

Stand in front of the projector so it projects onto you, not the screen, thus distorting the slide.

Clip art. Lots of generic clip art.

Make your organization disagree with your topics. Say, "I have three extremely crucial points" and then list two or four, so people will wonder if you forgot something somewhere. And go into detail on three of the four, but not the fourth one.

Find yourself very, very funny. Use humor that apparently only you get (cliched or not), and be very proud of that.
posted by Madamina at 1:05 PM on January 19, 2011 [2 favorites]

Put your critical images or text in dark blue on a black background or in yellow on a white background.
posted by pombe at 1:05 PM on January 19, 2011

Present an extremely complex diagram and talk through the whole thing rapidly as if recapping basics. Make sure to use an obscenely bright green laser pointer as you are doing so to make furious, gigantic circles that may somehow relate to part of the diagram you're talking about.
posted by solipsophistocracy at 1:05 PM on January 19, 2011 [2 favorites]

Make sure you include a random animation and sound effect AND make sure you go backwards through your slides once to refer back to something you mentioned earlier.

realize you "forgot" to include something extremely technical and crucial an attempt to ad lib it without any visual aids

depending on what all tech you're using you can also draw on your slides and forget to clear the ink

you can ask someone else to drive your powerpoint - preferably someone who doesn't know anything about your presentation, so you have to say next slide every time it's time to move on
posted by dadici at 1:06 PM on January 19, 2011 [2 favorites]

Awesomely bad science = stuff that blows up that maybe wasn't supposed to. Work that in somehow and you're golden.
posted by TooFewShoes at 1:06 PM on January 19, 2011

Uh ok, so I failed reading comprehension and went off on a rant. My bad. More constructive:

-- Depending on your audience, it might be appropriate to pretend to have a slide that was given to you by someone else, whose contents you don't really understand. (I've seen this happen with people giving talks that summarize a whole group's work.)

-- Don't have a laser pointer to start with. When you find one / someone gives you one, flail it wildly across your screen so it's hard to tell what you're actually pointing at.

-- If someone asks you a question, either (a) get horribly confused, (b) defensively avoid answering, or (c) give a meandering non-answer instead of asking them to clarify.
posted by dorque at 1:06 PM on January 19, 2011

Wacky sound effects! Lots of wacky sound effects! (And yes, I draw this suggestion from personal experience. *car crash* *sheep bleat*)
posted by Faint of Butt at 1:09 PM on January 19, 2011 [2 favorites]

Also, at some point you should accidentally go out of slideshow mode and struggle with the powerpoint interface to get it back (ideally waiting for someone from the crowd to yell instructions).

Bonus points if the minimize reveals something thoroughly inappropriate.

* Apologise for every second slide. "Sorry about the graph on this - I'll explain it."
* Promise to explain something in more detail later. Fail to do so.
posted by zamboni at 1:12 PM on January 19, 2011 [3 favorites]

Also, you could show up late - "Is this the room?" - and spend time putting the slides onto the laptop from a thumb drive.
posted by carter at 1:12 PM on January 19, 2011 [1 favorite]

Mix units.

Use the same variable, say X, for about three different things.

Use well known variables, like c or h for other things.

Include a poor scan of very complex equation.
posted by advicepig at 1:13 PM on January 19, 2011 [1 favorite]

Pretty please share the presentation when you finish it.
posted by verdeyen at 1:13 PM on January 19, 2011 [6 favorites]

Fidget - a lot. I'm a fidgeter and I consistently got low presentation marks because of this. Best fidgeting tool - one of those telescoping metal pointers.
posted by muddgirl at 1:13 PM on January 19, 2011

Give a nice ramble-y introduction which covers a bunch of irrelevant stuff in an incoherent and boring manner. Nothing like making the audience's eyes glaze over before you even get to the meat of your topic. Done right you can give a bunch of 'background' while still not explaining anything actually useful to your project, so the audience still don't know why you did your experiments etc (as Knowyournuts has suggested). Randomly introducing a brand new topic right in the middle when everyone is already bored is also fun for increasing the 'oh shit this will never end' factor.

Make sure your data has a lot of excruciating detail which you go over. Don't summarise anything. My favourite ones are epidemiological or genetic association studies where they give big tables full of data about odds ratios and stuff where they read out all the (many!) numbers but don't bother to say which ones were significant or give any kind of conclusion or take home story. The kind of thing which can be summed up in two sentences but instead is strung out into five, tedious minutes. Definitely make sure to put in some more subtle stuff like misleading graphs and diagrams too, as that's the kind of mistake that someone who thinks they're good is still likely to do.

Then when you run over your time, as any bad presentation is bound to do, refuse to stop. Say things like "I'll just quickly go over this one thing" then start back in with the tedious detail covered in a monotonous voice. Get slightly panicked about the rush, possibly even a little belligerent.

And lastly, vary your fonts up. I saw one talk where every sentance was a different font style, size and colour. The speaker read everything verbatim (so we really couldn't help but look) and our heads were spinning by the end by the sheer tastelessness of it all.
posted by shelleycat at 1:14 PM on January 19, 2011 [1 favorite]

If the slides will be running from a laptop, schedule your antivirus software ahead of time to start checking in the middle of the presentation. Schedule iTunes and Windows to look for updates. In short, have new windows popping up and interrupting left and right.
posted by sarling at 1:14 PM on January 19, 2011 [22 favorites]

Present serious data, such as death counts, with a little airplane piloted by an owl flying across the slide pulling a banner. I think Powerpoint 2003 has that one built in.
posted by ChrisHartley at 1:17 PM on January 19, 2011 [24 favorites]

Worst presentation pet peeves arising from actual talks i've seen:
1. If you have exit the slide presentation to go open some other software to display the movie or whatever that you can't get to embed or play the way you want to in ppt etc. Doubly irritating if it's not already open in the background so you have to go rooting around on your desktop (have some very unprofessional desktop background photo too!) to find it and open it and wait for it to launch. Do not continue your talk while this rigamorole goes on, but narrate the navigation of your files for the audience.
2. Eat while lecturing.
3. If you're a dude, wear disturbingly tight trousers. It will fascinate and disgust the audience.
4. Draw all arrows and cartoons in ms paint or something equally stupid. make all lines wavy and in headachy colors. Also make sure that none of the actual conclusive data is presented in raw form, only cartoons.
5. If your goal is to annoy, you can never be too patronizing. It really stokes the nerd rage of the audience into an inferno of hate, they will go for your jugular.
posted by Cold Lurkey at 1:19 PM on January 19, 2011

When you find one / someone gives you one, flail it wildly across your screen so it's hard to tell what you're actually pointing at.

Oh yes! And when you do point at something on screen don't point right at it, circle the dot erratically and quickly around the general region of the thing in question. Not just a few times but continually for several minutes as you talk about the thing. I don't know why people do that but it's actually pretty common (and kind of dizzying). The best kind of pointer is one that also has a button to make the slides change, so you can accidentally make them flip back and forth at inopportune times (which I've actually done and is really distracting for both speaker and audience).

Having some kind of physical pointer for some of the talk is also fun because you can misjudge depth and accidentally whack the screen really hard then look stunned. Maybe plan to have the laser pointer stop working part way through so you can swap.

My other problem is walking up to the screen and pointing to stuff right there, even rubbing my hands over the screen. That really does not work for anyone.
posted by shelleycat at 1:21 PM on January 19, 2011 [1 favorite]

• Flip back and forth between slides repeatedly while you try to figure out what to say.
• Fiddle with anything around you -- pens, keys, laser pointer, the noisier the better.
• Don't label any of your axes, or more subtly, label them in bizarre, non-standard ways (like an x-axis reading "-(time)" that extends from right to left, or a y-axis that just reads "performance" with no units).
• Make sure fonts get "corrupted," preferably into Wingdings. Do not in any way acknowledge the fact that the slide is illegible. Bonus points: "As you can clearly see..."
• Use completely gratuitous slide transitions and animation (e.g., letters "pouring" off the screen or catching on fire).
• Have the "video" showing your most important results replaced by a gray box reading: "QuickTime(tm) and a Video decompressor are needed to see this picture"; make no attempt to play the video, show stills, or explain what you should be seeing.
• Use non-standard fonts at tiny/enormous font sizes.
• Use clashing, low-contrast color schemes like teal on puce or red on green.
• Wave vaguely at the slide with the laser pointer instead of pointing to specific places; don't bother stabilizing your arm when you do this.
• If you have slides with text on them, read the text verbatim to the audience.
• Talk about as many different projects as you can cram into the available slides, as opposed to picking one or two and developing them.
• If you get stuck somewhere, bail out by saying something slightly condescending, like "and the rest of this should be obvious." Bonus points: "This should be obvious to anyone who's taken high school [computer science/biology/physics]." This is also a great way to answer questions. "Trivial" is another good escape hatch, as is "I don't have time to go through the math."
• Breeze through your actual results; belabor every detail of the technical procedures you used to get the results.
• Put every piece of data from your lab notebook into your talk.
• Play fast and loose with statistics. Call something with p = 0.2 a "trend," while not "formally statistically significant." Bonus points: elsewhere in the talk, use p > 0.1 as evidence against a trend.
• Use lots of jargon-y metaphors from different disciplines, especially when you need to obscure the fact that you actually mean something simple. For example, if something approaches a particular state, call that state an "attractor." Under no circumstances explain what an attractor is.
• Try to work in a good 20 seconds of awkward silence. Even better if you can do it in the middle of a sentence.
• Run over by a lot, but skip the acknowledgements due to time.

posted by en forme de poire at 1:22 PM on January 19, 2011 [4 favorites]

Show a video that's related to your work, but in a different language (e.g. Japanese). Make sure the audio is good and loud. Then talk over it, louder. It's absolutely maddening.
posted by xil at 1:23 PM on January 19, 2011 [4 favorites]

This presentation features a number of helpful terribly-designed slides (I know, I know, you already have enough good terrible ideas on that front, but I figure this question is crying for that video...)

More-helpfully... Have your talk be completely reliant on embedded video. Make sure that every video DOES NOT PLAY. Then, spend about 5 minutes *describing* the videos, preferably with weird wavy-hand gestures or little jumpings-up-and-down.

Take frequent long drinks of water. Wear a lapel mic. Make your drinking and swallowing noises really loud and distracting and slurpy. Then, completely lose track of your train of thought after every drink of water.

Have a cold. Sniffle a lot.

Shoot the laser pointer into the eyes of your audience every so often -- it'll wake 'em up!

Try to have a really distracting personal mannerism/nervous tic. The best one I've seen was the woman who (hopefully without realizing it) would grope her own right breast when perturbed. None of us had the balls to say anything to her.

Eat a snack -- preferably a messy or crunchy one -- while giving your talk.

Don't put in an important control. Get unreasonably angry when you are asked about it. Finally be all like "We didn't do it. We probably should." Then, have your first pocket slide appear after the acknowledgements. The first pocket slide will, of course, be of the important control experiment.
posted by kataclysm at 1:23 PM on January 19, 2011 [2 favorites]

Not sure if your topic is up for grabs, but I would think that a scientific presentation on the topic of something notoriously pseudoscientific (like, say Astrology), with plausible stats to "substantiate" its claims would be very illustrative.
posted by skechada at 1:24 PM on January 19, 2011

for emphasis, use bold, colour, AND underlining randomly combined (works best with comic sans). Include cheezy photographs or clip art to capture the 'emotional tone' (preferably more than one or even one of each type per slide). Make sure your hierarchy of headings is totally dysfunctional, contradictory, and mix your headings between numbers/letters chaotically.
posted by kch at 1:27 PM on January 19, 2011

Oh! List your agenda somewhere, so people know to expect X number of sections (let's say ten). Have "question time" very early on and go into lots of detail, or have lots of people ask questions, so the audience will wonder how in the world you'll have time for the other nine sections.

Go into detail about something that doesn't have anything to do with the presentation -- a funny story involving the researcher, or why you named your dog after Scientific Concept Y, or what happened when you were trying to collect the data.
posted by Madamina at 1:27 PM on January 19, 2011

Tell awful jokes relating to your chosen, highly obscure jargon. Switch media in the middle of a presentation. "My grad student will be here in a moment with the transparencies." Grad student comes in, you accidentally-on-purpose scatter the slippery pile of transparencies, then, as you stand up, stare, blinking, into the harsh white glare.

Ask for questions at inappropriate intervals, but then point at someone who has not asked a question and say, "What? I'm sorry, I didn't quite get that. No, I am almost certain you asked something."

Jargon jargon jargon. Discuss theory without evidence, draw conclusions without practical impact, relate utterly irrelevant anecdotes of the life of someone tangentially related to the field.
posted by adipocere at 1:27 PM on January 19, 2011

I was in a terrible mood when I started to read this thread, but the hilarity of some of these ideas have completely snapped me out of it. Please update us after the presentation. Video would be awesome.
posted by 8dot3 at 1:30 PM on January 19, 2011 [4 favorites]

Oh, yeah, I forgot about laser pointers.

Keep your laser pointer on throughout the entire talk, and always be circling something, relevant or not. That way, your audience won't be able to pay attention to what you are saying.

When you get questions from audience members, accidentally direct the laser pointer at each person as a way of calling on them. (No, don't. That's dangerous.)
posted by Knowyournuts at 1:31 PM on January 19, 2011

You could pull a Kary Mullis and have a slide with some nudity.
posted by exogenous at 1:32 PM on January 19, 2011

dress up and wear a wig with fuzzy white hair that stands up
posted by anniecat at 1:32 PM on January 19, 2011

-Skip or flip through slides. Particularly skip the important looking chunks saying "I'll come back to these".

-Don't finish. When time is up look dumfounded and say "while I guess my time is up". Alternately say "well I guess my time is up" and then carry on for some time still flipping frantically through slides.

-Start the presentation by saying "usually I give this talk in 2-4-6 hours but I've been asked to get it down to 20 minutes". Make it apparent that you have not edited the presentation to fit the new time slot but intend to skip around and/or talk faster.

-Start a presentation by saying "I don't really know much about this topic so I spent a lot of time last night putting this together with wikipedia" (I sat through this presentation and it was on organic beer brewing -- in a room full of beer brewers.) Include helpful commentary about things you just learned that seem important. (ie, different beers are brewed with different grains! this seems important!)

-Don't turn off your phone and fumble around when it rings (and/or answer it). Even better -- attempt to ignore it. When it rings a second time answer and/or turn off.
posted by countrymod at 1:32 PM on January 19, 2011 [1 favorite]

like, dress up in a lab coat
posted by anniecat at 1:33 PM on January 19, 2011 [1 favorite]

Make sure to make handouts of your slides to give to everyone. Make sure to make them 6 on a page or worse, to make your illegible graphs and small fonts even worse and totally unreadable.

(This may or may not be a real story from my job)
posted by utsutsu at 1:33 PM on January 19, 2011

The time-wasting slides are my favorite, things that eat up 5 minutes before the talk begins.

Acknowedgements with rigorous completeness. Be sure to name your funding organizations in full (the time for confusing acronyms comes when you're talking about the science). A list of everyone who contributed to your project, first and last names and what group in what institution. Read them all. Stumble over pronunciations. Thank people for "valuable discussions" but don't say what. Never mention their names in the talk to indicate what you might actually have had help with or done yourself. And don't forget to thank the organizers for inviting you, if appropriate for your mock-up.

While an outline can be a good thing, I've seen too many that don't actually include any information, for example:
Thursday Seminar:
1. Introduction (point and say "yeah, I guess we just did this")
2. Outline (especially good when this is underlined, to point out that yes, now we are looking at the outline, yet this is the only time you ever show a mapping slide)
3. Experiment (nothing about it, just "experiment")
4. Conclusions
posted by aimedwander at 1:34 PM on January 19, 2011 [2 favorites]

Talk in a monotone at anything other than your audience. The blackboard, your notes, the floor. Read exactly the text on your slides, including all the variables in any equations.

Use color poorly (especially good if your use of color would make a graph/equation completely incomprehensible to the colorblind).

Those bad graphs: don't label the axes, and have no idea what the labels should be. Be confused about the units, and have at least one feature of the graph that you can't explain (what is that dip about? I don't know.)

At least once, answer a question by trying to go to another slide-- but hunt around for it for a good long time.

Get off on a tangent with one particular audience member when no one else knows what you're talking about. (hard to do in 20 minutes, but maybe you can think of a way)

Find a podium to hide entirely behind. Especially good if you're not very tall; if you *are* tall, extremely bad posture can have a similar effect.

Have a weak red laser pointer and move it around/turn it off so fast your audience can't figure out what you're pointing at.

Use very long sentences with multiple clauses; refer to variables, units, concepts, equations, etc that you haven't bothered to define; introduce vastly more notation than you can possibly use in 20 minutes.

If your field (or that of any of your students) ever uses blackboards, exhibit terrible board use-- write equations in random order, or use a very tiny portion of the board and just keep erasing the same spot. Have awful handwriting when you do this. Bonus points if you can find a place to write where more than half the audience can't see it.

Dress way out of character for the situation-- depending on the field in question, a full suit or sweatpants could be worse. This isn't rational, but some people will get tripped up if your appearance is too far off norm for the field (and this is totally field and situation dependent. In my world, you can really only screw up by being dressed too formally, and there are a few types of presentation where that's not possible. But YMMV.)

Insult audience members- either for the questions they ask, for not asking them, for breathing funny, whatever. A belligerent attitude is a great block for information transfer. Alternatively, be overly deferent; whenever an audience member questions you, agree with them, even if it contradicts what you just said or what you're about to say.
posted by nat at 1:36 PM on January 19, 2011 [1 favorite]

Two weeks ago I gave this exact assignment to four TAs so they could generate the worst design reviews possible for engineering undergrads to pick apart. The short list was:

* Extraneous details
* Splitting information between far-apart designs
* Long sentences that say nothing. "Combining these two designs offers both reduced cost and an opportunity to leverage existing GEC competence to recreate the "industry standard" at significantly lower in-house cost."
* Explaining the wrong thing (eg, the physics of baseball laces instead of how many outs in an inning)
* Gambling words: simple, rudimentary, clearly, obviously
* Acronyms w/ no precursor
* Slides with horrible backgrounds (one TA found a glorious printed circuit board image that did a good job of obliterating almost every word on her slides).
* 99% of slides spent recapitulating the problem statement/thesis/specification that everybody already knows instead of conveying what your actual contribution was.
posted by range at 1:39 PM on January 19, 2011 [3 favorites]

Plant accomplices in the audience to ask ridiculous questions. When you answer them, say "That's a great question!" a lot.
posted by zamboni at 1:40 PM on January 19, 2011 [3 favorites]

auto-start video
posted by ennui.bz at 1:42 PM on January 19, 2011

A few minutes into the talk, "accidentally" smear some chalk on your face while scratching your nose. (Or dry erase marker, if that's what's available.) The few times I've seen lecturers do this, it's so supremely distracting.
posted by enlarged to show texture at 1:44 PM on January 19, 2011

You need to slam open the door, say something ambiguous in meaning, but very possibly representing a terrible personal event, and just LAUNCH into all other presentation errors without stopping to explain what you meant. Just keep going, and ignore the look of horror/concern on faces in the audience. I suggest...and I'm not saying I'm drawing from experience here, but here are a few ideas:

'Sorry! I'm running behind! My husband/wife didn't come home last night!'

'Sorry! Things are a bit delayed just now. If I seem to be having trouble with my arm, it's because I was hit by a car!'

'I had to sleep in the horse stall last night!'
posted by Uniformitarianism Now! at 1:46 PM on January 19, 2011 [4 favorites]

This might be inspirational.
posted by timsteil at 1:47 PM on January 19, 2011 [2 favorites]

I wish I could find this again, but many years ago on USENET, I saw an exchange which went something like:

Person one: "Wester's Dictionary defines (something) as ..." is the worst possible start to a presentation ever.

Person two: "No, the worst possible start to a presentation is actually: 'Hello. Today I talk about non-linear transformations in fourth order non-differentiable Bessel space. Please to excuse in advance my not-so-very-good English ..."
posted by Comrade_robot at 1:47 PM on January 19, 2011

To second namewithoutwords http://www.youtube.com/watch?v=i3RqDp7i308
posted by elroyel1327 at 1:49 PM on January 19, 2011

(Sorry, I realize I should have been more specific: after you get the chalk/marker/pen on your face, continue lecturing, blithely unaware of the fact that you now have, say, an asymmetrical chalkdust mustache.)
posted by enlarged to show texture at 1:49 PM on January 19, 2011

Try to have a really distracting personal mannerism/nervous tic. The best one I've seen was the woman who (hopefully without realizing it) would grope her own right breast when perturbed. None of us had the balls to say anything to her.

Similarly to this, if you by chance have a prosthetic or assistive body part (false limb, glass eye, even a hearing aid or eyeglasses) contrive to remove/misplace it during the presentation.
posted by Faint of Butt at 1:52 PM on January 19, 2011

Go off on a tangent about your personal theories of the origins of the cosmos and quantum physics.
posted by Burhanistan at 1:52 PM on January 19, 2011

Assuming that your audience will have a printed copy of your slides, digress off one of the first slides and ramble a bit, covering several future slides-worth of information in your digression. Then go back to the script and cover it all again in the monotone.

If you can have a confederate interrupt with a question or two that prompt these digressions, all the better.
posted by chazlarson at 1:55 PM on January 19, 2011

Oh, and if you have printed slides, make sure they don't match your actual presentation ever so slightly.
posted by MustardTent at 1:58 PM on January 19, 2011

Please use a poorly-scanned, comic to highlight one of your slides - perhaps a nice Dilbert or a vintage Ziggy. Even better if the comic isn't really that funny, or doesn't really relate to the topic.

Draw something really important on the white board that you forgot to include in your slides. But make sure you use a marker with barely any ink left in it and a worn-down tip. Erase part of it with your hand, and then get ink all over yourself and clothes. Set the marker aside so you don't use it again, but then realize you have to use it because it's the only marker in the room. Narrate each step with mild cuss words, like "oh crud" or "fudge".
posted by SuperSquirrel at 2:00 PM on January 19, 2011

If people ask for clarification during the talk say "I'll address that during the question period at the end." Then be sure to go over time and wind up with "unfortunately, we've run out of time for questions..."
posted by MsMolly at 2:03 PM on January 19, 2011

Make sure to make handouts of your slides to give to everyone. Make sure to make them 6 on a page or worse, to make your illegible graphs and small fonts even worse and totally unreadable.

Also, the slides in the handout should be in a different order than the actual ones you're using in your presentation, with some of the presentation slides entirely missing from the handout, and other slides that appear in the handout but not the presentation. Bonus points for numbering them and referring to earlier slides by the wrong number.
posted by DevilsAdvocate at 2:04 PM on January 19, 2011

I agree with using an unnecessary slide transition between every slide. The longer it takes, the better. Sound effects? Awesome. Even better if you couple it with going back to search for a slide, so you get to see all the transitions again, but in fast motion, making them look even stupider than they did the first time. Also agree with the clip art and especially bad text-on-background color choices. Any combination of red, blue and yellow is generally going to hurt to look at after a few minutes. Have lots of really light pictures, and don't turn the lights off in the room until you've gone through several of them. Realize it, turn off the lights, and backtrack.

Trying to explain a complicated video that you couldn't get to work: yes.

My favorite talks were always the ones with gratuitous name-dropping and bragging about irrelevant stuff. ("This experiment was from when I was in XYZ's lab at ABC institution during the time when they were inventing the cure for cancer. NBD.") Actually, it's even better if the name dropped is for something really uninteresting ("I did this while working with QRS, you know who he is, he discovered the 237th splice pattern of this gene. AND I HELPED. nbd.") Nothing better than hearing for someone go on for ten minutes about the great institution they were at and all the cool stuff that happened there but they were not involved in. The less relevant to the topic of the presentation, the better. You could do this to replace the intro/ background part of the talk, which is great because it doesn't prepare the audience at all for what you're actually going to talk about AND it bores them and makes them think you're a tool.

I think using acronyms over and over without explaining them should DEFINITELY be included. That happens all the time! Especially if they are kinda crappy ones that couldn't be said out loud without sounding stupid, like ZGARMOF. Or that sound kind of weird or dirty like PEEMAT.
posted by GastrocNemesis at 2:05 PM on January 19, 2011 [1 favorite]

- Start off with a terrible clearly dumb joke. I suggest knock-knock.
- Have a lot of papers that cascade to the ground are which are out of order.
- Adjust your bra straps a lot and wear something that is staticky and clingy and very very distracting
- Attach the microphone right next to your jangly necklace [alternate: have no place to attach a microphone, also have no idea how to use a microphone]
- Bring water/coffee with you and then misplace it, spend time looking for it
- Say um and you know a lot, too much
- Have a slide with an obvious error in it and look at it like you've never seen it before
- Pace like a caged panther
posted by jessamyn at 2:10 PM on January 19, 2011 [2 favorites]

Get into an in-jokey discussion with the most important person in the room. Refer obliquely to shared experiences with that person often during the course of your talk. The more irrelevant, the better!
posted by MsMolly at 2:11 PM on January 19, 2011 [1 favorite]

- Stare at your slide notes. Read directly from them.
- Make sure the slide notes do not fully correlate to the slide content (as if you're reading from a different draft than you're presenting). Alternately, the only words out of your mouth should be exactly what the audience is viewing.
- When not staring at the slide notes, stare at the ceiling. Do anything besides look at the audience. Especially when taking questions.
- Anything you say in discussion should deflate the presentation's assertions, as you'd done further analysis between preparing the slides and doing the presentation.
posted by ardgedee at 2:12 PM on January 19, 2011

Leave your mobile phone on loud. Have someone call you during the talk. When it rings, wonder aloud who it could be, pick it up and then look at the caller ID. Then tell everyone that it's your proctologist/criminal defense attorney/other inappropriate-to-mention professional.

Also a true story.
posted by grouse at 2:12 PM on January 19, 2011 [2 favorites]

Microsoft powerpoint gives you lots of options: slide animations, colour schemes, sound effects. You need to use all of them or the software engineers will feel bad. (You are using powerpoint, right?).

Bonus points: prepare you presentation on a Mac, but display it on a PC. Ensure your mental purity by never ever trying it on a PC before the conference.

Read all text, exactly as displayed, in a monotone. Don't waste extra words explaining anything.

Make certain that there are at least 250 words on each screen. You want to be certain that the audience reads you paper. Reading it for them is therefore double-good.

Go through experimental methods or theoretical derivations in detail. If it was hard to do, it should be explained in detail. Preferably this should be a standard method specified by regulation or a textbook derivation known to any second-year undergraduate.

In tables, all text should be no grater than 12 point. Using an 8 point serif font is preferred.

When inserting graphs, use low-res black and white scans. You want to ensure that your graphs are nice and blocky. Be sure to smudge the photocopies before scanning them.

Don't bother attributing other people's work, especially figures. If your audience is current in your discipline they will figure it out for themselves.

Plan on 1 slide every 15 seconds, then add a few more to be certain. For a 20 minute talk, have at least 100 slides prepared.

Speaking of which, moderators and session chairs are only there to offer suggestions. You don't need to take their frantic gestures seriously. They've probably got some unfortunate syndrome like Tourettes or something. Best to just ignore their social difficulties.

Make certain to use at least one, if not more fonts that are installed only on your machine. this will ensure a nice boxy presentation at the conference.

Because you want your talk to be light-hearted and up beat, you should do your main text in Comic Sans. Alternately, if you want to be serious, use Times Roman, but at sizes no bigger than 12 points. A serious talk should be seriously hard to read.

Include multimedia in some instrument- or camera-specific codec that MS Windows does not have installed by default. Bonus points: make certain that Windows Media Player has never been initialized on the conference computer for maximum fun. Offer to install your weird codec at the podium, but use a pirated codec pack that is a virus payload. Make certain that all links to movies or sounds are hard coded to the directory structure on your home machine.

Similarly, be certain to include some live web content in your presentation. Use a custom app on an underpowered server located in rural Alaska.

You don't need to practice you talk at all. You know your subject. You don't need to rehash it in your room.

Set your screen saver to 1 minute or less. Be certain that it is password-protected.

Master Class:

Go old school: prepare your talk on viewgraph acetates with multi-coloured markers. Hand draw your graphs. Store them without interleaving paper so that they rub off on each other. Use old acetates that haven't been properly cleaned.

Give a colleagues paper. After declaring that you don't know much about it, start criticizing the work shortly after beginning.
posted by bonehead at 2:15 PM on January 19, 2011 [7 favorites]

This would be pretty much an entirely different approach to a terrible presentation, but if this is to grad students it might help address a relatively common problem:

Get up with a slide A CURE FOR CANCER or something like that, and then one slide that might be almost interesting... then spend the next 18 minutes talking about how someone had cancer and it's terrible, and why you first became interested in cancer research, and a horrendous literature review. But the important thing about each piece you bring up is what it inspired in you, etc. At the end, when you have 2 minutes left, have the chair or a co-conspirator tell you your time's almost up and then rush through a confusing mess of acronyms and charts with lots of arrows until you're cut off in mid-sentence.

Acronyms: use them inconsistently. Be sure to pronounce big, unpronounceable ones (ie ones that are worse the SCMODS), but spell out Enn Ay Ess Ay or refer to the NAS Administration.
posted by ROU_Xenophobe at 2:16 PM on January 19, 2011 [3 favorites]

Throw in a political barb during your talk on an entirely apolitical topic. ("This reaction achieves a higher rate of consumption of the arsenate ion than has ever been observed before, much like Obama with your income tax money.")
posted by DevilsAdvocate at 2:17 PM on January 19, 2011 [2 favorites]

@ grouse "soon to be ex wife" was my favorite from one of my professors . . . he would constantly get calls/ texts during class and make horribly uncomfortable, jokey comments about her. He also wore horrible tight jeans that you could see his package in. Girls would go out of their way to avoid sharing an elevator with this guy.

So yeah, pretty much any personal, TMI offhand comment to make the audience uncomfortable is great.
posted by GastrocNemesis at 2:19 PM on January 19, 2011

Use hand-written or -drawn material whenever possible despite your skills being terrible and the results barely readable. No such thing as overdoing this one.

Also, depending on your acting skills, it might be fun to respond to your own mishaps and failures with an air of suppressed panic ("KEEP IT TOGETHER - MUST MAINTAIN PROFESSIONALISM - SALVAGE SALVAGE SALVAGE") and resolute refusal to acknowledge that anything is going wrong.
posted by jinjo at 2:23 PM on January 19, 2011

Any presentation can be enhanced through the use of hand-outs. For note-taking I suggest printing at least a dozen slides per page so you need a magnifying glass to read them. You can also hand out 'supplemental materials'- basically photocopy some tangentially-related-out-of-date-and-possibly-out-of-print text book. It helps if they are photocopied poorly, crooked, blurry, too dark or light... and you want at least a phone book's worth of paper involved.

For bonus points have some other peripheral to pass around the room, but make sure it's not anything actually interesting. I'm thinking something along the lines of a plain glass beaker, a chunk of limestone, or a normal chicken egg.
posted by Green Eyed Monster at 2:25 PM on January 19, 2011 [1 favorite]

grouse has reminded me: wear a pager. Get paged. Loudly. Call them back in the middle of the talk. Don't remove your mike--just leave the room. Take a few minutes to chat about a case. In graphic detail. For example, if you're lecturing to people outside of your field, those who work outside of the health professions love hearing about eye injuries, so let them listen in. Please include phrases such as 'So the globe has ruptured, and the contents are completely extruded?'

This is especially good if your current position is in basic science and research, and you yourself are not involved in clinical practice. Because that's the kind of thing that will just stir up an audience while they try to figure out what the hell is going on.

It doesn't even matter what's on your slides after that. It could be all unicorns and rainbows for all they'll notice.
posted by Uniformitarianism Now! at 2:52 PM on January 19, 2011 [6 favorites]

Take a science paper (or just something not summarized) and copy and paste it into PP. Not a single bullet, picture, graph, etc. Just a giant wall of text and read exactly like Ben Stein would in Ferris Bueller's Day Off.

Other than that, I would use a certain idea mentioned in the thread for each slide. One is full of flames, letters dropping, airplanes, etc other crap. Another has aforementioned pointless and convoluted graphs. Another with one single word and then just ramble on about that subject and how it has to do with [insert some polarizing topic here].
posted by jmd82 at 3:18 PM on January 19, 2011

On at least one slide, dense with text, in addition to left-to-right, some text is bottom-to-top, some top-to-bottom, and some of the text is upside down. Then mirror one block of text. If you can do this on a graph, so much the better.
posted by exphysicist345 at 3:20 PM on January 19, 2011

Ask someone to show you how to advance the slides and work the laser pointer.

Have some mildly-embarrassing picture as your desktop background.

Have a slide in there twice.

Be passive-aggressive and defensive. E.g., "as that SMART guy in the back probably already guessed, here's what we found."

Comment on how few people are there because they probably thought another session was more interesting and make everyone physically get up and move closer "so you won't try to sneak out in the middle too."

Definitely, the video that can't play.

Color choices backwards: bright red for the mundane, pale yellow for the most urgent, another pale indistinguishable yellow for a much less urgent condition.

Mention some discovery that "we" made that everyone in the room knows is someone else's work.

Have the title slide have last month's conference title on it.
posted by salvia at 3:24 PM on January 19, 2011

Horribly unintelligible accent, combined with mumbling such that even with clear speech you couldn't be understood.

Highlight this by having nothing actually legible on the slides. Go over to the board and pick up a writing utensil, giving the audience some hope that you'll actually write SOMEthing they'll be able to understand. Raise the utensil to the board, pause, then turn back to the audience instead and wave the utensil around in aimless circles while continuing to speak unintelligibly.

Eventually, have something written on a slide that is text. Contradict it with a few suddenly clear words in the middle of the mumble--for example, if the slide has "NaCl" on it, say "Niacin."

Have a few clearly, egregiously wrong bits in the slides, something that everyone would know is completely wrong (like "E=mc^4" or "H2O + NaCl -> CaCO3" or "Live birth of ostrich larvae"). Just go on by these as if you don't realize how wrong they are.
posted by galadriel at 3:31 PM on January 19, 2011 [1 favorite]

Wear very large lens glasses that bounce light so no one can find your gaze.

Stuff your hands in your pockets, look at the floor, drop stuff.

Get some of the information almost correct, or use common misconceptions and support them.
posted by effluvia at 3:40 PM on January 19, 2011

A large, animated gif that has no conceivable link to the subject semi-obscuring the text. Make reference to it along the lines of "I thought you'd appreciate the picture, hahaahaha!"
posted by Coobeastie at 4:00 PM on January 19, 2011

If it's a 20 minute talk, talk for 40 minutes, and then entertain any and all questions. Bonus points if lunch was supposed to be at the end of the 20 minute talk.
posted by Wordwoman at 4:11 PM on January 19, 2011

Draw some graphs in MSPaint.
posted by advicepig at 4:13 PM on January 19, 2011

Put your c.v. on several slides at the beginning of the talk and spend a few minutes explaining to your audience why you are the definitive authority on your subject. List other recent invited talks you've given, including those to your parole board, your mother's book club, and your nephew's third-grade science class.
posted by tully_monster at 4:16 PM on January 19, 2011

Have a complex but tiny graph on 1 mm graph paper drawn in felt tip. Scan it. Print it. Scan it again. Lower the contrast a bit. OK, now it's ready to be pasted into your slide.

Plan on 1 slide every 15 seconds, then add a few more to be certain. For a 20 minute talk, have at least 100 slides prepared.

Personally, I do this all the time, but I loves me some build slides (and have seen to many people impaled on their own animation software) so 10 of my slides may convey the idea of one of someone else's.

That being said, do a build slide WRONG! Don't start with the final image and then peel things off one by one - build it fresh each time so as you move through the progression the image shifts and jumps around. Redraw you text box in a different place on each slide and use a different font -OR- keep adding to the same text box and letting your software keep shrinking your text to fit, so that by the last slide your audience is trying to read eight point type and has motion sickness.
posted by Kid Charlemagne at 4:29 PM on January 19, 2011

Use pi and e as variables.
posted by Kid Charlemagne at 4:32 PM on January 19, 2011

If you have control over the projector placement, definitely make sure it has a bad keystone shape due to misalignment. Preferably so bad that one edge is out of focus.

Put long blocks of text and then read them verbatim.

Make every sentence on the slide a complete sentence including articles. Then, make each one appear with a animated transition with sound. Preferably a long one.

You MUST put in that one slide that's just a bad, blown-up to grainy pixellated, watermarked photo of a scantily clad woman, then make the "how'd that get in there? Well, now that you're paying attention..." joke.
posted by ctmf at 4:44 PM on January 19, 2011 [2 favorites]

Have several slides that are just large tables, at least 10x10. Make sure all the p values <>
Always include cartoons found at images.google.com that are heavily watermarked to indicate that they are copyrighted, but "Reproduction rights are available from cartoonstock.com"

For your penultimate slide show your "bibliography" for the talk. Make it hundreds of references long in 8 point times new roman.

The final slide should be blank except for something in giant 96 point (preferably animated) 3-D Word Art type font. Good choices are either "THANK YOU!!!!" or "??QUESTIONS??".
posted by roofus at 5:00 PM on January 19, 2011

Also make sure to make your presentation "interactive" by simply turning all your statements into fill-in-the-blank questions, quiz show style. Especially sentences which directly relate to your key point, which you haven't told anyone yet, so they have no chance of knowing the answer.

Break your presentation up into two power point files for no reason whatsoever, so the audience has to watch you on the projector trying to find the second one ten directory folders deep. Click wrong a lot.

Anything that will distract the audience from what you're saying right now. My favorite: advance to a really fascinating-looking chart, or a really detailed diagram. Then say, "ok, before we go on, a bit more about the last point..." but DON'T BLANK THE SCREEN OR BACK UP THE PRESENTATION. Just leave it up and talk a few more minutes while everyone tries to concentrate on you and not look at the tempting distraction up there.

Draw a piping diagram or electrical schematic using MS Paint freehand.

Pick one verbal tic and say it over and over. I like the "sling blade Carl" mmm-hmm after every point.

We actually used to have to rotate who was doing weekly training presentations for the group at work. Power-point required, even for topics clearly not suited for it, because someone had decided that a blanket "power-point is a good training tool and we need to use it" policy was a great idea. We'd do exactly what you're asking on purpose all the time just for humor.
posted by ctmf at 5:00 PM on January 19, 2011

Pronounce acronyms wrong. SQL should be "squall." If you have equations with greek letters, read them as "squiggly thing." (If there are two, the other one is "other squiggly thing")
posted by ctmf at 5:11 PM on January 19, 2011

Start five minutes late because you can't figure out how to set up the projector.

Pick a meaningless modifier to overuse as filler when you speak. "Basically," "actually," and "definitely" are good. If you can stuff a couple flagrantly misused "literally"s in there, even better.

Start out with a slow, in-depth explanation of something very basic that your audience already knows. Transition to something they don't know and might have trouble grasping, and breeze through it assuming they'll keep up.

Include a Rickroll. Act like you discovered Rickrolls yesterday and they're the funniest thing in the world and you just have to share with everyone. Extra credit if the video takes a really long time to load, or if you call him "Rick Ashley."
posted by Metroid Baby at 5:13 PM on January 19, 2011

The things that piss me off the most about bad presentation are the following:

1. Not knowing the audience intelligence level. You don't need to explain gravity to a bunch of Physics PhD students, but you can't show complex statistics to a bunch of elementary school children and expect them to understand what it means. So: Alternatively give too much and too little detail and exposition. Preferably front-load the presentation with extremely detailed information about past work.

2. Not doing any sort of analysis. Try desperately to avoid explaining why your topic is interesting or worthwhile. If there are statistics, make sure that they have minimal real-world meaning and cannot reasonably be visualized by the majority of your audience. Don't make any conclusions.

Let us take, for example, the major earthquakes in China that killed thousands through buildings that were not adequately built to withstand earthquakes. To make this into a presentation which would drive me mad, your presentation would go as so:

Introduction--1 slide overview about earthquakes in China. Put up numbers about strength of earthquake in the scale least understandable by the audience--ground movement as a percentage of g might be good. How many buildings collapsed? Possibly say how many people died as an afterthought. Spend less than 30 seconds on this slide.

Move into historical building methods of China. Have about 10 slides on this, with mostly text and only one or two images, generally blurry or shot through trees or otherwise unsatisfying. Clip art of rocks and logs also acceptable. Discuss in detail building materials and styles. Don't provide reasoning as to why they chose particular building materials. Possibly go into detail as to why they chose particular stylistic features like curvy roof corners. Once you get through the building methods of ancient China, spend a slide saying "But they recently adopted modern Western building techniques" but don't explain what those techniques are. Act like they are clearly superior to Chinese building techniques.

Move on to a picture of a collapsed building from the earthquake without any intro. Gruesome would certainly make me cringe. Then immediately move on to detailed information about specific stress fractures and collapse types in the buildings with piles of charts. Make sure not to point out anything interesting. Are these types of problems common in earthquakes? Are there certain building techniques that made some types of problems unusually common? Don't let your audience know, just sound really excited about the data. Then talk about one particular type of break found that is perhaps an unusual one but don't let anyone know that it was unusual. Give a slide or two of equations, a couple slides of close-up pictures or bad drawings of the types of breaks.

Then finish. Make no conclusions, or extremely banal ones. "Total building collapses caused many deaths in China" "Earthquakes cause stresses on buildings that are not found in normal conditions"

Make sure you don't know the answers to any questions not explicitly stated in your presentation. If someone tries to make you analyze your results, hem and haw and just restate the data. Possibly state completely new data but don't relate it back to the presentation at all. Or say "But we did do this as well (state something seemingly significant) but it wasn't related so we didn't include it"
posted by that girl at 5:22 PM on January 19, 2011 [1 favorite]

In addition to not explaining the complicated math, it might be fun to go into extensive detail about simple math. Don't worry about defining the Ricci tensor or what einstein summation is on one slide. On a later slide, though, go into extensive detail on the definition of "polynomial", "variable", and "plus sign."

Present nearly redundant data in multiple graphs on multiple slides. Flip through the slides quickly (30 seconds for 10 slides), since the data is redundant.

Only idiots need labels on their axes. Your audience isn't full of idiots, is it?

Do you have a pet? I'm sure your audience wouldn't mind if your cat slept in the front row.
posted by bessel functions seem unnecessarily complicated at 5:42 PM on January 19, 2011

Lots of good things here. The only thing I have to add is that you should ensure that your display settings are wrong so that some portion of your slide isn't projected at all. Put the important stuff in that part of the slide.
posted by juliapangolin at 6:01 PM on January 19, 2011

- write things on the board semi-illegibly using crumbly white chalk
- mumble about things different from what you are writing on the board
- never ever face the students
- ask questions but go right into answering them yourself

Based on a real professor.
posted by meepmeow at 6:05 PM on January 19, 2011

Employ all of the above AND make your presentation about how to give an awesome presentation.
posted by guybrush_threepwood at 6:25 PM on January 19, 2011 [2 favorites]

Wear a t-shirt with unrelated, distracting, and unprofessional text on it. Something moderately offensive, like this. (Unless you work with minors.)
posted by desjardins at 6:35 PM on January 19, 2011

Mis-button your shirt. If you are female, wear a red bra under a white shirt, or wear knee-highs under a skirt that is not QUITE long enough. If you are male, leave your fly open, or wear white athletic socks.

Have every single slide be backed by an eye-popping gradient fill, like from magenta to lime green. Make sure every slide has a DIFFERENT color combo. Have the text be in a color that contrasts eyebleedingly with one color and barely stands out from the other. The last slide should be rainbow.

When you set up your laptop for the presentation, make sure your laptop wallpaper is something inappropriate. Not taters, just a frat boy in mid-vomit or a genital-referencing LOLcat or something along those lines.

Leave one slide up while you give the material for the next four slides. Realize your mistake, and just hop past all those slides while muttering without giving your audience a chance to process the information.

Have at least one slide that appears to be related, but spend twenty seconds staring at it and then confess "I have no idea why that's in there."
posted by KathrynT at 6:37 PM on January 19, 2011

Clear your throat a lot. Chew gum or noisily unwrap hard candy.

Worst thing ever - spit tobacco in a cup. (yes, I went to school in Montana, why do you ask?)
posted by desjardins at 6:42 PM on January 19, 2011

This is very fun and I wish I could attend, but... what's the point? It might be better if you have a "Before/After" structure to contrast presentations that are agonizing with those that are great.

Examples:
- The opening: Make all the technology mistakes you can (can't find slides, can't make projector work, etc.). Contrast with preparation tips.
- Presenting: Do all the bad presenter ideas in the thread above. Contrast with good speaking tips.
- Technical slides: Again, look to the above then show well-formed slides and explain why they work.
- Transitions.... you get the idea.

Again, while I think this is a fun idea, listening to 20 minutes of this will be, well, boring and frustrating unless you provide a constructive takeaway. And "don't do any of this" isn't enough of one.
posted by sfkiddo at 6:57 PM on January 19, 2011

See Professor Irwin Corey
posted by Drasher at 7:04 PM on January 19, 2011

Don't run it in presentation mode. Just move to each slide as if you were editing it in powerpoint.
posted by milestogo at 7:09 PM on January 19, 2011

Good grief, I leave you people alone for six hours and look what you do! These are *amazing* and I am going to have so much fun with this. I'll be going through to mark best answers, which are mostly geared toward my particular audience (undergrads), though for a presentation seminar for practicing scientists some of the more black-belt level tips/disasters would be fantastic. The presentation's not for a few weeks, so feel free to keep the awfulness coming in!

For sfkiddo and other wondering about context etc. (which I realized after posting I forgot to include), I'm a grad TA for a methods course. I will give the Worst Presentation Ever and then (ala range's comment) the students will critique, and then they will get an actual tutorial on presentations from the professor.


Some examples of things I intend to do:

* Have far to many slides and rush through the last 10 or so in two minutes

One more suggestion of my own, after rereading of my post: No proofreading whatsoever. Definitely not.

Don't know if I can my supervisor to go for videorecording, but I'll make my slides and notes available once I'm finished...
posted by heyforfour at 7:53 PM on January 19, 2011 [1 favorite]

As an add-on to the part about copying a slide from another presentation, have all or some of that slide be in a language other than the language of instruction.
posted by mskyle at 7:55 PM on January 19, 2011

Ensure that stacked-bar graphs have patterns that are both eye-popping and close to indistinguishable, and that color fills blend in with the background so it just looks like error bars floating in the middle of a pair of axes. Use error bars only some of the time.

Line graphs need twenty or more lines of slightly different colors going in all directions. Assert, "Each of these is an individual subject (or rat or country or study site or whatever), but you can see the overall trend in the data is really obvious". Stick a text box on the graph that says "p=0.05032". State "As you can see, it is statistically significant." Under no circumstances specify what "it" could mean.

Tables must be more than 5 x 5 cells, and must have illegible header rows and footnotes. Always use as many decimal places as possible, especially for p-values and for things where you've noted you only have very imprecise measurement.
posted by gingerest at 7:57 PM on January 19, 2011

If anyone walks in late, or tries to leave early, be sure to publicly scold them for it, right in the middle of your talk. (Arrange to have an accomplice do this if you like.)
posted by DevilsAdvocate at 8:03 PM on January 19, 2011

Always use as many decimal places as possible

especially for reporting uncertainties and variances. Be sure to report at least two more significant figures than your smallest uncertainty. Feel free to randomly use expanded uncertainty, relative and absolute standard deviations as the mood takes you.
posted by bonehead at 8:42 PM on January 19, 2011 [1 favorite]

Use "we" to describe only a subset of the work in the talk "when we did this experiment, we found..." only don't have that at all be related to who did the work. Bonus if "we" did the work before you were actually born.
posted by lab.beetle at 9:12 PM on January 19, 2011

Try to wedge into your introduction what Meyers-Briggs type you are. Then, as an aside, tediously explain what all the letters mean. I've seen that three times and I vividly remember wanting to claw my face off each time like it just happened even though it was years ago. Bonus points for appearing to have just heard of it and calling it "new."
posted by ctmf at 10:20 PM on January 19, 2011 [2 favorites]

All the real key skills for a terrible presentation have already been well covered, but here are some extra bonuses:

(Repeating this one for emphasis though it's already been mentioned.) Read the entire presentation quietly and quickly off the page, absolutely never looking up. Bonus points for consistently putting your emphasis on the wrong words and/or only pausing for breath in the middle of very long sentences.

Mention at the beginning (or in the middle, in an impromptu aside) how long it took you to prepare the presentation. A grossly inappropriate length of time in either direction is good — for a very complex piece of the only research project you've been working on for the last decade, say confidently that it took "nearly an hour" to assemble the slides, as though this meant the presentation was ironclad and perfect. But for a very simple piece of research (big bonus if it seems like it's 101-level material but also the only research project you've ever done in your life!) you want to uncomfortably reveal that you've spent the last 3-6 months honing the presentation, complete with numerous dry runs. Combine with the following tip for maximum audience discomfort:

Talk up front (embarrassedly but well-rehearsedly) about how you're shy, but you're working hard on your people skills. Mention Toastmasters or some equivalent public-speaking practice and talk about how much it's helped your self-presentation. Make sure you still seem very uncomfortable and shy the whole time, and that the only sign of your people-skills practice is a handful of off-topic jokes randomly injected into the middle of the presentation (maybe some really ridiculous slide transitions, too). When you totally flub the presentation at this point, everyone will be cringing sympathetically as well as in annoyance.
posted by RogerB at 10:55 PM on January 19, 2011

While you're struggling to figure out how to get the slide show to run, 'discover' a new PowerPoint feature you didn't know about, such as the ability to draw on slides while the slideshow is running. Spend several minutes gleefully playing with this feature. For bonus points, draw something wildly inappropriate.

Liberally pepper the speech with your personal views. I was once treated to a (for credit!) presentation on the Sarbanes-Oxley Act, during which the student went off on a three minute rant about how much better society would be if we used the death penalty to deal with financial fraud.
posted by PercyByssheShelley at 1:10 AM on January 20, 2011 [1 favorite]

Copy and paste a large chunk of text from the internet, but make sure you don't edit out any hyperlinks, and they remain on your slide as underlined blue text.
posted by chronic sublime at 2:24 AM on January 20, 2011

As an add-on to the part about copying a slide from another presentation, have all or some of that slide be in a language other than the language of instruction.

Even more fiendish than that. Try to find out if someone in your audience is fluent in a language other than the language the presentation is in. Works best if your field is truly international, of course. Have a few slides in that language. Then say, "My [ForeignLanguage] skills aren't so great, but this means incomprehensible grammatically incorrect blather, isn't that right, [ForeignLanguageSpeakerInAudience]? No? Could you please be so kind as to translate for us? It would be a shame for such important information to go unknown. I'll pay you later! Ha Ha Ha!!!"

Then move on to another slide with the same problem. "Since [ForeignLanguageSpeakerInAudience] has been so kind as to translate before, would you please again? You're so helpful, gosh I'm glad you're here."

Third slide. Same thing.

I spent 3 days of a communication course, supposed to be in English, where precisely this happened. I was the surprise-designated interpreter, being the only native English speaker bilingual in French. The other students did not speak the instructor's native language (French). No compensation whatsoever — that's where the truly fiendish part comes in. You say "thanks", "gosh it's valuable", "I'm definitely buying you lunch!" in this gushing, grateful tone, and then... that's it. Not even lunch. I totally let him have it on the third day. Especially when he couldn't figure out how to say how important it was in communication to take audience language into account in order to avoid one-sided scenarios.
posted by fraula at 2:27 AM on January 20, 2011

Realize that you've forgotten some crucial bit of information, and then write it on the whiteboard directly underneath the projected image. It ends up being both completely illegible and highly distracting. Bonus points if you have to pull up the projector screen to write on the board, and then leave the screen at half mast for the rest of the presentation.
posted by daniel striped tiger at 2:38 AM on January 20, 2011

Make it clear that you've never before seen some of the slides you're presenting. Pause, preferably with a deer-in-the-headlights look, and make little guttural noises while you try to figure out how to present this never-before-seen slide (which exhibits all the horrible behaviors listed above--dense, unorganized, illegible). Ideally, there's completely unrelated clip-art featured prominently.

You could steal a page from the SciGen guys and have one slide consist solely of a picture of Che Guevara. Upon that slide appearing, deadpan "And this slide shows that our approach is truly revolutionary."
posted by Mayor West at 4:57 AM on January 20, 2011

One thing I'll add is that we were surprised when we actually did this with undergrads - they were WAY less cynical than we expected and took the presentations much more literally than we planned for. After getting bit by this on day 1, we consciously dialed it back on day 2 and it went much better.
posted by range at 5:09 AM on January 20, 2011

Spend far too long hunting through the menus of the software trying to make it go fullscreen. Elicit advice on this from the audience but frustratingly fail to carry it out.

Have the slide subtitle contradict the rest of the contents of the slide.

Assert something surprising and complicated, and leave the proof 'as an exercise'.

Repeatedly use one word when it's obvious you mean a different one.

Ask the audience an incredibly vague question and then be disappointed when nobody gives the answer you wanted.

Ensure your desktop is a confidence-sapping mess of distractingly interesting/incriminating icons.

Draw a diagram on the whiteboard but get it completely wrong, erase it and start again.

Pick a saying or idiom and repeatedly misuse it.
posted by d11 at 5:35 AM on January 20, 2011

One more thought - you're already putting WAY too much info on each slide - make sure you include a couple of slides with one (preferably tangential) thought in headline size across the middle - with no additional info
posted by dadici at 8:11 AM on January 20, 2011

Read every word of every slide, and occasionally repeat extremely simple things for emphasis. EXCEPT for one complicated and potentially interesting slide which you breeze by in about 3 seconds saying, "and here's a graph."
posted by dirtdirt at 9:50 AM on January 20, 2011

I just did this last night in class. :-( Ask the audience a question without realizing the answer is right there on the slide in front of them. Act all weirded out when nobody responds to your question, because they're all "Duh! It's right there on the screen, you idiot!" Turn around and read the slide, and say "Oh, it's right there on the slide, isn't it." Wonder why everyone is rolling their eyes at you. Whimper a little.
posted by SuperSquirrel at 10:07 AM on January 20, 2011

use animations to transition between EVERYTHING.

Star wipes. Nothing but star wipes.
posted by LordSludge at 10:11 AM on January 20, 2011 [2 favorites]

Sent this to my dad (who's got years of experience in this, poor thing), who added:

HA!.

I LIKE it.

I don’t know why, but they missed: Bounce on every word with a red laser pointer as you read the slide. Then wiggle it around and circle things really fast creating a little laser trail that burns into the audience’s retinas and makes everyone feel a little queasy.

posted by fiercecupcake at 1:05 PM on January 20, 2011 [1 favorite]

We have a terrible presentation work. Its hallmark is it would do stuff like this:

HOW TO EDIT A PAPER:

* How to edit a paper

* Now we will discuss paper editing

NEXT SLIDE >

He would then read it all, verbatim, including the repeated phrases.
posted by chairface at 1:09 PM on January 20, 2011

Copy and paste a large chunk of text from the internet, but make sure you don't edit out any hyperlinks, and they remain on your slide as underlined blue text.

Even better, copy from wikipedia and leave the 'citation needed' notation. Then later you get to talk about both plagiarism and why wikipedia is not a reputable source of information.
posted by shelleycat at 1:27 PM on January 20, 2011

Imitate Star Wars Episode I, and don't put your main character anywhere in the first half.
posted by talldean at 1:55 PM on January 20, 2011

Be sure to laugh very loudly at your own jokes, especially if they're not very funny.

Add a video clip from a movie only loosely related to the point you're making. Fail to edit it correctly, so it goes longer than necessary or takes longer than usual to get to the point. Apologize several times to the audience during the excess time, explaining you're not sure how it ended up like that.
posted by telophase at 2:18 PM on January 20, 2011

The one that always kills me is when the talk uses a really inappropriate "funny" image taken off the Internet to illustrate a serious point. Thus you can throw in a tried and true tacky jpg like (and I wish I could say these were hypothetical examples) a naked obese man or a toothless Rastafarian to illustrate some scientific conclusion.
posted by girl scientist at 2:28 PM on January 20, 2011

The worst presentations I've seen (or given) have been have been ones given with slides designed for a different audience and a longer time slot, so that the presenter has to decide which content is not important on the fly. Nothing irritates me more than someone skipping through a few slides saying "you don't need to know this", or similar, as it clearly demonstrates they are just recycling some old talk.
posted by larkery at 3:29 AM on January 21, 2011 [1 favorite]

Structure the presentation as if you were summarising a report. Use headlines introduction, theory, method, result...

Develop some kind of bizarre habit. For example stare sideways into the left wall the whole presentation. put your glasses on top of your head and bring them down with a quick head jerk repeatedly or stare intensely on on person in the audience until they look back, then you quickly shy away.

Tell everybody why this subject does not really interest you and what subject you would really like to be teaching.
posted by furisto at 6:52 AM on January 21, 2011

By multiple request, here are the fruits of your labors/comments.

Sadly for me, the professor had laptop/dongle issues on the first day of class, so I ended up skipping the can't-load-my-slides portion. It's still pretty terrible, though apparently I did far too good a job explaining what they would have seen in my nonworking video :) Also, I wore my loudest and clackiest shoes only to discover I'd forgotten the classroom is carpeted. Sadness.

Here's a video of the presentation (my phone goes off at around 10:05.) Here's the slides in case you need a closer look at how bad they are.

The paper I based the talk off of is actually really good, though dense. I delivered it more or less backwards.
posted by heyforfour at 11:58 AM on February 8, 2011 [21 favorites]

Thanks for posting this-- it is amazing on so many levels. The parts where you refer to portions of the slide obscured by your shadow are hilarious. Also how you keep jingling your keys. & you really pull off the "image not loading" confused look.

A triumph!
posted by activitystory at 10:44 AM on February 9, 2011

Amazingly, the slides make my copy of Open Office crash in a way I've never seen before. Very exciting. Thanks for the links.
posted by jessamyn at 11:01 AM on February 9, 2011

I'm late but:
Tell them what you are going to tell them, tell them, and then tell them what you told them.
posted by thinkpiece at 11:10 AM on February 9, 2011

At some point, blame somebody else for your level of disorganization.
posted by philip-random at 2:27 PM on February 9, 2011

Have the organizer of the conference deliberately pretend the typeset research paper is your prepared and edited slide presentation and show only the former to the audience. (This actually happened. Knowing it would, I brought overhead-projection printouts and carried on.)
posted by joeclark at 12:53 PM on February 26, 2011

Another winner: Post your slide video in Windows Media format, which nobody on anything other than Windows can watch.
posted by joeclark at 12:55 PM on February 26, 2011

Coming in late, but having an IM window pop up with semi-personal questions is always fun.
posted by bigenchilada at 8:36 PM on March 12, 2011

I'm getting a 404 error on the links, perhaps they could go on the wiki?
posted by Blasdelb at 6:51 PM on September 29, 2011

VeryBadPresentation.pptx (1.08 mb). That will work for a time.
posted by cashman at 6:50 AM on September 30, 2011

omg, the fish background. Amazing.
posted by en forme de poire at 7:16 PM on October 14, 2011

Wow. My favorite was Slide 10, but then I saw the multicolored letters on Slide 17 and the absolutely unreadable bouncing text of slide 18. Genius.
posted by salvia at 2:47 AM on October 15, 2011

I finally watched this. Absolutely brilliant. Good show.
posted by grouse at 8:02 AM on October 15, 2011