Is Baseball the Least Random Sport?

Former front-office analyst and now stats professor at Smith College Ben Baumer has a paper out, with cowriters Michael J. Lopez of Skidmore College and Gregory J. Matthews of Loyola University Chicago, that hopes to answer a question we’ve all thought about when our favorite team loses: how often does the best team win in a given sport? How much of our pain can we explain away with luck? The answers contain multitudes.

Their paper, which clearly attacks the main problems in answering the question, can be difficult for a layman to understand. But the basic genius in their approach may be using the betting line to judge team strength. That number allows them to compare different sports with different scoring systems and different amounts of team games.

After testing the line and finding it was a good estimate of team strength separately in each league, the analysts had to figure out the home-field advantage in each sport, the parity between teams, and then the estimated season-to-season and week-to-week variability.

Guess what: baseball had the smallest home-field advantage, with every baseball park scoring below every other sports’ locations other than the Rockies, who have a bigger home-field advantage than four hockey teams. Baseball also had the most parity and the smallest variability week to week and season to season.

So you might have guessed their conclusion — baseball is, indeed, the sport with the most amount of luck involved. The teams are packed in the closest, they have the smallest advantages from their parks, and the so the games are closest to coin flips among the sports. (Note: per Baumer et al., CDF denotes “cumulative distribution function… of 1000 simulated game-level probabilities in each league.”)

This, of course, has ramifications for running a team, ramifications we’ve probably seen recently. The combination of the added Wild Cards and all this parity has led to teams diving for the middle. The Cubs and Dodgers aside, most teams seem content to build a team with a true-talent win total in the mid-80s and hope the balls bounce their way in a given year.

You’ve seen evidence here of these sorts of things. When Dave Cameron wrote that losing on purpose was a bad idea because team projections can easily be off by 10 or more wins, when Billy Beane says “sometimes good things happen,” when the Braves sign Sean Rodriguez — that’s when you know that baseball itself knows how closely packed in the teams are.

But what does this mean for us as fans? That’s less clear, because luck means different things in different situations. When it comes to losses, luck absolves our team of guilt. And though parity and loss absolution are probably good things for the average fan — your team has the chance to win in any given game, and if it loses, you don’t have to feel so bad — it does break down on the extremes.

If every game was truly a coin toss, we wouldn’t want to watch, would we? There would be no meaning. We’d be watching the plinko chips bounce their way down to some random destination. Front offices would have no reason to build good teams, talent would be randomly distributed and randomly expressed, I’d guess. We’d be watching blips on a screen.

The absurdity of that language probably gives us a clue that we aren’t anywhere near that point in baseball. The Cubs had a great team and they won a bunch of games. They still lost 60-plus games, and that’s great. That’s a testament to the game’s parity — best in sports, apparently.





With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.

40 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Zonkmember
7 years ago

I always felt like the NBA was the least random. I mean, each year you can probably pick from 4-5 teams who the champ is going to be, with a strong probability attached to just 2 or 3 (Warriors, Cavs). In the NBA, you can make the playoffs and still have virtually no hope of winning a title; not true in baseball

That’s why baseball is great

bawfuls
7 years ago
Reply to  Zonk

Agreed the NBA is least random, but there are benefits to that as well as drawbacks. The NBA Finals and/or Conference Finals are nearly always a showdown between the best teams in the league. We almost never get to see the best two teams in baseball face off in the World Series anymore.

Adam C
7 years ago
Reply to  Zonk

Having 87-88 win teams win the World Series is not great. It’s crappy actually. The NBA has the best of both worlds; a long playoff season and still get the best teams in the NBA Finals. Since the NBA went to the 16 team playoff format in 1984 only one team seeded lower than #3 has won the NBA Finals. The 1994-95 Houston Rockets were the #6 seed but they were the defending NBA champions.

Anon
7 years ago
Reply to  Adam C

Yep. Just to piggy-back on what you’re saying – since 1984, those 94-95 Rockets were the only NBA team to win fewer than 54 games and won the title. THey only won 47 but had an inner-circle HOFer (Olajuwon) and traded mid-season for another HOFer (Drexler) and as you noted, were the defending champs. But it gets even better than that re: the NBA. I haven’t checked it in a few years, but as of a few years ago no team since 1984 had started a season worse than 6-4 in their 1st 10 games and won the title.

The NBA really needs to either shorten the season a bunch (I’ve seen suggestions of home-and-home with every other team – 58 games) or go to 8 team playoffs. Neither of which will ever, ever happen of course, so this post is pointless but still – the NBA season is wayyy too long. Then the playoffs are also wayyyy too long.

Jon L.member
7 years ago
Reply to  Zonk

Season length vs. single-game reliability:
http://pages.ucsd.edu/~nchristenfeld/Publications_files/Sports.pdf