Unexpected scores in sports
May 26, 2009 7:47 AM   Subscribe

Which of the major US sports (baseball, football, basketball, hockey, soccer) has the most variance in final score?

I am curious to know which sport has the most variance in final score. I am a math cretin so I may not be asking the right question, but hopefully you geniuses can surmise what it is I am looking for.

For example, the average score of a baseball game is about 5-4. However, its not abnormal to have scores of 12-8, or 1-0. Football and basketball seem to have a variety of scores, too. Anedoctally, soccer seems to have a very slight variance in score, as most professional games seem to end with a small amount of scoring by both teams, from 0 to 4 goals. As a sub-question, I guess I'm trying to identify what score should induce a "wow" from a fan... ie, is it more unique to see a soccer game that's 5-3 or a football game that's 53-35?

I hope that's enough explanation of the question. Thanks
posted by RajahKing to Sports, Hobbies, & Recreation (17 answers total)
 
Well, if you reduce that 53-35 football game into single point scores, you'd end up with something around 7-5 (assuming extra points were made in each case). And that's a lot less impressive looking (IMO) than 53-35.

My point is, it's not necessarily the variance in points that gets fans excited, but the higher number points. In baseball, for example, a 19-18 game would wow a lot of people because there's a lot of action. But if you add variance along with high scoring, say 19-2, it's the ultimate wow because you get lots of action and a blowout.

But it also depends on who you're asking, because some baseball fans will be much more wowed at a 1-0, 10-inning game than a 19-2 blowout.
posted by nitsuj at 7:55 AM on May 26, 2009


My point is, it's not necessarily the variance in points that gets fans excited, but the higher number points.

I don't know about that. Every basketball game seems to be 103-101. I often wonder why they don't just count the shots they miss.

And yes, a 1-0 baseball game is much more tense and exciting than a 19-18. A 19-18 would be a laugher, entertaining for the sheer ridiculousness of it, but nobody would call it a "good game" in a million years.
posted by rokusan at 8:05 AM on May 26, 2009


Football is my hunch, if only because of the big 7+ point leaps you can get by scoring a few touchdowns.
posted by rokusan at 8:06 AM on May 26, 2009


i think a better way of going at this would be to look at the outliers in each league and see which leagues have less parity. So, NBA and MLB, you have some really good teams, then there is a big drop off to the crappy teams, and not much in the middle. With the NFL, there is more parity due to salary cap, and therefore, it's literally any given sunday, any team can win, most of the time.

So, my suggestion is looking to see what leagues have the most blow-outs and you'll see leagues that have less parity.

My vote would be NFL, there are a few bottom dwellers, that have had little luck doing well, but every season there are surprises.
posted by fozzie33 at 8:08 AM on May 26, 2009


I guess I'm trying to identify what score should induce a "wow" from a fan... ie, is it more unique to see a soccer game that's 5-3 or a football game that's 53-35?

With (American) football in particular there are some lower scores that are rare enough that they might have a "wow" factor. 7 (touchdown + extra point) and 3 (field goal) are the two most common types of scores, and 6 (touchdown without an extra score) and 8 (touchdown + 2-point conversion) are also common enough most higher point totals can be reached without looking strange. But a score like 5-4 would have to be an execptionally strange game, because one team would have needed to score two safeties in a single game (the record in the history of the NFL is three) and the other team would need to have had a safety and a field goal.
posted by burnmp3s at 8:08 AM on May 26, 2009


I don't know about that. Every basketball game seems to be 103-101. I often wonder why they don't just count the shots they miss.

Right... but that's not considered a high-scoring game in basketball. 186-184 (Pistons vs Detroit, 1983) would be considered a high-scoring game, and no doubt wowed a lot of people.

A 19-18 would be a laugher, entertaining for the sheer ridiculousness of it, but nobody would call it a "good game" in a million years.

Really? Maybe you wouldn't... but think of how many people would walk away from the stadium during a 19-18 game with 10 home runs thinking, "wow, what a great game!" What they're really saying is, "wow, I had a great time, that was fun!" Hardcore baseball fans will, of course, enjoy the closer games, but casual fans (e.g. everyone else)? A "great game" is one with a lot of action in it.
posted by nitsuj at 8:22 AM on May 26, 2009


(Pistons Denver vs Detroit, 1983)

Ack.
posted by nitsuj at 8:23 AM on May 26, 2009


But a score like 5-4 would have to be an execptionally strange game

Yea, I think it's because teams score mostly in 7's and 3's that some scores are very rare. A season or two ago, some game was like 17-11 or an equally strange total, and the commentators were commenting on how it'd be the first game to finish with that score in like 40 years. As has been mentioned, the 7-point touchdowns can quickly put a team up by 20+ points, even if it was actually close the entire game. If an NBA game wins by 20+ pts, chances are it was never close.
posted by jmd82 at 8:47 AM on May 26, 2009


Every basketball game seems to be 103-101. I often wonder why they don't just count the shots they miss.

Because then no one would ever take a shot.
posted by dfan at 8:59 AM on May 26, 2009


Best answer: This article computes the variance in runs scored / game for individual MLB teams over the course of a season. They seem to range from about 8 to 14 (so standard deviation is going to be in the 3-4 range.)

It's not clear that "variance" is really the right measure for sports like baseball and soccer, where the distribution is highly non-symmetric thanks to a lower bound reasonably near the mean. It might be most appropriate to think about the variance of the _difference_ between the winning and losing score, which I'd think would be closer to a normal distribution (though this is of course an empirical question.) If you really want to think about the distribution of the total score for both teams, you might want to take into account correlation between the two scores -- relatively low in baseball, very high in basketball (where the overall pace of the game pushes both team's scores in the same direction.) So that it's not so weird to see an 8-1 score in baseball, but I think it would be more unusual to see a 111-70 score in basketball (even though the individual team's scores, and the total score, are all within normal ranges.) This correlation also explains why the "Pythagorean constant" is much higher for basketball than it is in baseball.
posted by escabeche at 9:01 AM on May 26, 2009


Consider, too, that the number of points isn't always as revealing as the way in which they were scored (in football, at least.) 14-6 looks like a pretty normal NFL score, but if the six points were scored as the result of three safeties instead of two field goals or a TD with no PAT, that would be very wow indeed.

I suppose baseball could experience some degree of this too. 5-4 is a normal enough score, but if those five runs were scored on bases-loaded walks, and the losing pitcher also threw a no-hitter, that would be pretty wowing as well.

In summary: I like where you're coming from, but I really think there are too many variables in this to make any sort of meaningful statement about it.
posted by SpiffyRob at 10:58 AM on May 26, 2009


Study the spreads used in Las Vegas sports books for a year or two. Whichever sport has the highest average spread is the sport with the most variance in final score.
posted by wabashbdw at 11:17 AM on May 26, 2009


FWIW, if you ask people for the greatest games in any sport, I expect most would name close games with last-second finishes.

I'm a fan of a major college football team and an NFL team. Both have won big games and championships, but I'm often drawn more to the close wins (and pained more by the close losses) than the 45-13 blowouts when it comes to watching highlights on TV, DVD, or YouTube.

Then again, if you're in your home stadium and your team is blowing out the other team, it can be the best three hours of your life.
posted by Flying Saucer at 11:23 AM on May 26, 2009


In hockey, the score isn't always indicative of the "wow" from fans.
Most real hockey fans I know like a game with a lot of good scoring chances, physical contact, and excellent saves by the goaltender.

The goals are nice but consistent scoring chances and plays are what usually make for a memorable game.
posted by zephyr_words at 11:41 AM on May 26, 2009


Response by poster: To clarify: I am looking for the regularity of "outlier" scores, I think, and trying to figure out which sport you would "expect" more outlier scores from. More of a mathematical question than a qualitative assessment of good, exciting games.

"Wow" as in "that's a score you never see" as opposed to "wow, that was a great, close game".

Thanks, interesting stuff so far.
posted by RajahKing at 12:24 PM on May 26, 2009


My hunch is that soccer has the least variance. In the premiership, where teams are quite closely matched, a large majority of games end 0-0, 0-1, 1-2, or 0-2. The rare games where the margin of victory is >=3 goals, or any game with 8 or more goals are both pretty damn thrilling.
posted by roofus at 2:16 PM on May 26, 2009


I feel like the dependent nature of scoring in baseball must factor in somehow. In most sports, individual scoring events are independent. Scoring a goal resets things and the next goal must start from scratch. But in baseball, it takes a lot of effort to score one run, but then it is easier to score additional runs.
posted by smackfu at 10:49 AM on May 28, 2009


« Older How to go from friends to more   |   Atheism definition aka Philosophy 101 Newer »
This thread is closed to new comments.