There's been some talk in the past about whether the playoffs are basically a coin flip (as I contend) or not (as others contend). I just finished updating my SDMB OOTP history spreadsheet to include everything up to the 2017 season. Here are some numbers from SDMB OOTP history that I found interesting about playoff performance.
In the 25 playoff series where one team won more regular season games than the others, the team with more regular season wins won only 11 of the 25 series and went 65-73 overall (.471).
In the 22 playoff series where one team had a better Pythagorean record, the team with the better Pythagorean record won 11 of the 22 series and went exactly 62-62 (.500).
By way of contrast, the home team has gone 75-65 (.535) in the playoff games so far.
I'm sticking with my coin flip theory.
You can use your fancy spreadsheet if you want, Mack, but all I know is that I've played you what seems like a bazillion times in the playoffs in two different leagues and you've beaten me every single time. If it's truly a coin toss, then that is one hell of a coin you've got there, buddy!
ReplyDeleteConcur with Mack.
ReplyDeleteI actually wrote a paper for The Baseball Prospectus many years ago with just this sort of analysis. I compared the historical playoff performance back through the 1940s. I used games won, team OPS and team ERA to compare playoff teams head to head.
In all but one case I found that the outcome of a playoff series was within a few points of random. That is, team performance and statistics were not an indicator of victory in either a five or seven game playoff series.
The one exception? I found that if a team was significantly better than their opponent in ALL three factors (20% better or more) then their likelihood of winning went to about 10%. So a team that's clearly dominant would win about 60% of the time. Otherwise: random.
A short series just allows for too much luck.
Matt, I'm pretty sure you won either one or two series against me at the start of the OEL. I'm definitely not unbeaten against you in the playoffs.
ReplyDeleteBy the way, if anyone wants this beast of an Excel spreadsheet, just drop me an email and I'll make it available via Dropbox.
ReplyDeleteOh sure, maybe I beat you back when I had even less of an idea how to play this game. Now that would be ironic. :)
ReplyDeleteInteresting theories.
ReplyDeleteI have absolutely no evidence except anecdotal, but I wonder if a team's record in the second half of the season and September influences how it will do in the post season. I refer to, as I often do, my incredible 2005 Stockholm team, which lost only three games after September 1st, including none to Mack's Seattle team in the Cecil Cup. (Yes, once upon a time there was a team that repeatedly dominated Seattle in the Cecil Cup playoffs.)
That Stockholm team caught lightning in a bottle and probably would've won ten straight against Mack's team if it was a best of 19 series. I recall that Seattle had a slightly better regular season record than Stockholm that season, too.
I don't disagree with the luck/skill theories above, but I think that there might be another reason for the above results.
ReplyDeleteI think that most of the time, that the teams are very evenly mateched in the playoffs this leads to the seemingly random results. I would be interested to see how far the variation is in the OPS and ERA numbers that were involved in the statistical analysis that you did Nate. If they were close enough then it may not matter.
But for sure you can look at Saskatoon winning 2 seasons ago, or San Fran winning in MLB this year or Arizona and Florida over the Yankees years ago. There is a getting 'hot' aspect that makes things seem random that is for sure.