Drop and give me 100!

June 23, 2008 9:43 AM

My friends and I are about to embark on a 6-week push-up training program. Obviously, this has to be a competitive event. The problem, though, is finding a fair way to measure who wins...

I came across Hundred Pushups last week and, being terribly out of shape, decided to do it. After telling some fellow type-A friends, we're turning it into a competition, but running into difficulty figuring out how to measure. All of the identified options have drawbacks, so we're looking for new suggestions or ways to modify any of the items below:

(1) Who can do the most pushups after 6 weeks - Some people are more in-shape now than others, so this would be unfair towards those starting behind
(2) Absolute increase in pushups - Requires a "weigh-in" at the beginning, where people can cheat (consciously or subconsciously) and not do as many pushups as they're cable of. Also probably has starting-point issues like point (1).
(3) Percentage increase - Has the same weigh-in issues as (2) and also gives advantage to people who start weaker. E.g., someone who starts at 25 would have to get to 250 in order to compete with the guy who starts at 10 and gets to 100.
(4) Something based on BMI - I was hoping to find a chart of "If you're BMI is x, y pushups are expected" online, and measure success relative to the chart. No dice finding such a chart, though
(5) Everybody over 100 wins, everybody under 100 loses - same starting point issues at (1), nobody in the group likes ties

Thanks!

posted by um_maverick to health & fitness (14 comments total)

Go with #3. It's the only fair way, and, besides, why would your friends cheat you?
posted by mpls2 at 9:54 AM on June 23, 2008

How about absolute or percent increase in chest size? Not sure which is more fair, but at least you get rid of the doing-fewer-pushups-than-you-actually-can part of 2) and 3).
posted by Grither at 9:54 AM on June 23, 2008

I would say do 2 but add some sort of incentive for everyone to do well. For example, have the person who does the least amount of pushups pay some amount of money to the person who does the most. Or give everyone one "ticket" for each baseline pushup they do, and randomly draw a ticket out of the group to get some prize. That way there is a competition for everyone to do as many as possible when setting their baseline.
posted by burnmp3s at 10:03 AM on June 23, 2008

How about work output!

If you weight 200lbs, and can do 10 pushups that means 2000 lbs.
If your fit friend is 150 and can do 13 pushups, that means 1950 lbs.

YOU WIN!

Since there may be a subconscious or conscious desire to rig the baseline by your or your friends (if based on improvement), then go with work output.

OR you can do something totally crazy like:
Who can pushup 15,000lbs first.

The guy who weighs 150 lbs would have to do 100 pushups, the guy who weighs 200lbs would have to do 75 pushups. Time it. If the 150lber can do it in 6 minutes, and the 200 guy can do 75 pushups in 5 minutes, the 200lber wins. He did more work output first.

The best I could do was inversely relate skill to weight (the greater the weight, the lower the capacity to do pushups). This will totally work out...unless one of your friends is a bodybuilder. That would suck. Otherwise, I feel it should work.

Good luck...

Oh, by the way...I went from 12 pushups to 347 pushups one summer by doing as many as I can in a row during commercial breaks. Thanks TV!
posted by hal_c_on at 10:10 AM on June 23, 2008

Take a set of before and after pictures (shirtless, with faces excluded) for all involved. Then have a panel of neutral observers judge the best improvement. This kind of assessment is harder to cheat than 1 and 3. Assign a set point value for most improved and give decreasing rewards for everyone else. (So if there are 10 participants, the winner might get 10 pts., the next 9, etc., until you get to 1 pt. for least improved.)

Combine this with 1 and 2 above, giving the same points for first, second, third place, etc. .

The winner is the one with the best overall score. By combining two relative improvement scores with one absolute achievement score, you should be able to get a pretty good idea of who's done the best, while accommodating different levels of beginning fitness.
posted by oddman at 10:10 AM on June 23, 2008

I've been in charge of a weekly pushup contest in the past. There are freaks out there that weigh 60lbs and can do 200, and have arms like beanpoles. And there's me, who can't break 20... by just practicing once a week. When I ran my gig, I made sure there were as few barriers to participation as possible. Don't pair situps as well, don't add bells and whistles.

If you guys really want to encourage the best outcomes for everyone, then make it purely participatory. Whoever follows the schedule, honestly affirming that they max out when required, wins. When it gets down to it, the problem is not "not doing enough pushups", it's forgetting, or not finding the time to do it.

That, and oddman's picture idea.
posted by gensubuser at 10:24 AM on June 23, 2008

There is no way you will be able to do this competition without someone having an advantage. You won't be able to control everyone's form, and a 'weigh-in' is flawed because the weight distribution across each of your bodies will be different, along with your heights and arm lengths, all of which affect the actual amount of weight being moved during a push-up. Take that times one hundred and mister thunder-thighs and mister tyrannosaurus rex-arms will have a significant advantage.

That being said, I like number 5. It's quite an achievement to do 100 push-ups without stopping. And if nobody likes ties then I say you need to do something similar to hal_c_on's work output suggestion.

But you can't use body weight, you need to use the weight they will be pushing. So get a scale and have each guy get in the push-up position with hands on the scale to measure the weight each will be pushing. Also, I believe work is force * distance. So a proper assessment would involve measuring the distance each pushes himself as well. Get a tape measure and have each participant get in a push-up position and measure distance to the ground and then have each lower to a competition-certified position and measure the distance from the ground. Take the difference for the distance pushed.

Push-up Weight * Distance pushed * Reps = Work output.

You could also try this nifty Work & Power Output Calculator I found, though I'm not sure how accurate it is.

If you want to test improvement you'll still run into the problem of cheating on the initial test. You're all friends so an honor system should be okay. If not, then don't let a guy stop until his face is bright red and his arms are shaking.
posted by trueluk at 10:39 AM on June 23, 2008

ooh - I like the work output idea. I think we're going to do a percentage increase in work output as our metric, maybe with a bonus for the person who does the most overall (because that's still an achievement).

Thanks everybody!
posted by um_maverick at 10:51 AM on June 23, 2008

Actually, I'm a bit of a dope - while that seems like the perfect solution, the math works out exactly the same as in % change of push ups (assuming weight remains constant). I still like the work output idea, though....
posted by um_maverick at 10:53 AM on June 23, 2008

Get a old style "service" bell or something that dings or squeaks when the bottom is hit then your contestants will have less chance to fudge the bottom. The Range of Motion is where I can see some issues in measurement. Understood criteria for the ROM, pace and form should be decided before training and gaming.
posted by jade east at 10:59 AM on June 23, 2008

You're going at this the wrong way. We've got a active pushup group, and as best I can tell, the goal that's in a lot of people's minds is for us to just be able to do a ridiculous group pushup effort somewhere public. Kind of a pushup flash mob. Think more in terms of a big geeked-out spectacle instead of just having one winner!
posted by anildash at 11:42 AM on June 23, 2008

The problem is that you want to make it fair, which requires knowing TWO points of measurement reliably, but you can only guarantee the fairness of ONE point. That's no good. That's like an A/B comparison where A can equal anything.

This reminds me of an old riddle: Once there was a king with three sons. The king was dying and had to decide which son to leave the realm to, since he didn't want to split it up between them, but he couldn't decide which one. So he holds a competition--a horseback race. In order to account for the inevitable variation in his son's horsemanship skills, the race would be won by the horse that crossed the finish line last. But the sons could still cheat by simply riding extremely slow... So the king issued a two word instruction before the race began that would absolutely ensure a fair race. What were the two words? ANSWER (hover your mouse over the link).

How does this relate? Well, you need some way to ensure the baseline measurement is fair by making the contestants want to do as well as they can.

The answer is to split the contest into two halves: the winner of the combined results is the overall winner. In the first half of the contest (the initial baseline), each contestant tries to get in as many push-ups as they can in a certain amount of time... say, two minutes. Rank the results from lowest to highest (1 = lowest, # of contestants = highest).

For the second half of the contest, you have the same measurement as the first (get as many push-ups as possible in the same timeframe). Only for the results, you compare the number to the original and rank by ratio. Add the first half rank with the second half to get the overall winner. In the case of a tie, the person with the higher rank in the second round wins.

For example:

First half results (name/# of push-ups in 2 min.):

1. Bill (10)
2. Steve (20)
3. Hank (50)
4. Igor (60)

Second half results (name/# of push-ups in 2 min.):

• Steve (30)
• Bill (80)
• Hank (95)
• Igor (100)

2nd results/1st results ratios (ranked):

1. Steve (30/20)
2. Igor (100/60)
3. Hank (95/50)
4. Bill (80/10)

1st rank + 2nd rank (final results):

1. Hank: 3pts + 3pts = 6pts
2. Igor: 4pts + 2pts = 6pts
3. Bill: 1pt + 4pts = 5pts
4. Steve: 2pts + 1pt = 3pts

I tried to illustrate all the possible scenarios in this example. Clearly Bill is trying to influence the 2nd rank by under-performing for the first half, but you can see it doesn't help him. Hank and Igor both tie in points, but Hank wins the tie-breaker because he had a larger percentage increase.
posted by Civil_Disobedient at 8:40 PM on June 23, 2008

To be clear in case it wasn't obvious: the first half of the competition is done before you guys start working out, he second half is after the workout period.
posted by Civil_Disobedient at 8:43 PM on June 23, 2008

Hey all, just wanted to circle back and share the algorithm that we settled on - it's a weighted combination of most pushups at the beginning, # of pushups increased, and % increase, normalized, and weighted. It gets around the sandbagging of the initial "weigh in" by counting that towards the final prize, while also rewarding the growth as well. We did everything out in Excel, and it didn't transfer completely to google spreadsheets, but you can still get an idea of what it looks like with some fake data by clicking here

In that example, the person who went from 15 to 130 comes out as the winner, followed by 45 to 140, 25 to 140, and 40 to 140. Because each category is based on how much more you do than the worst person, sandbagging in any given category doesn't necessarily give you an advantage. Seems like it'll work...
posted by um_maverick at 10:59 AM on June 25, 2008