PDA

View Full Version : An SQD history of "momentum" . . .


Johnny Mostil
09-17-2005, 12:37 AM
. . . short, quick, and dirty, that is . . . and the short story is it doesn’t matter . . .

Given some of the posts I’ve seen about “momentum” recently, I wondered how “momentum” has affected teams in recent postseason play. I decided to test this with the results of the nine 162-game seasons that have featured three divisional winners and “wild cards” in postseason play from each league. That yields 72 observations (four teams in each of two leagues over nine seasons, or 4x2x9). I beg the indulgence of any sabermetricians among WSI’s 5,858 members (and counting!) for what I know is a flawed analysis (more on that below, but I’ll note here I haven’t done this type of work in a while, and have never before dabbled in sabermetrics specifically), and welcome suggestions and comments.


I decided to look at each playoff team’s record over its final 27 games (more on that number below as well) and the number of games it won in the postseason. If “momentum” is important, then we should expect teams winning more games in their final 27 to win more postseason games.

This doesn’t seem to be the case. I ran both a correlation and a linear equation and found statistically insignificant coefficients. In fact, the coefficients were negative—meaning teams winning more games in the final 27 actually lost more in the postseason—but, as noted, the numbers weren’t statistically different from zero. (And even if they were significant it still doesn’t mean it’s better to lose games at the end of the season, of course—just that there’s no evidence for the effects of “momentum” here.) I’d be glad to share the data and my calculations with anybody wanting to look at this further.

Non-statistical analysis of some teams also found no effects for “momentum.” Only once in the past nine years (2003) did a team having (or tying for) the most wins (among playoff teams in both leagues) in the final 27 games win the World Series. Only five times (Yankees in ’98, Braves in ’99, Yankees and Marlins in ’03, Red Sox in ’04) did a team having the most wins (among playoff teams) in the final 27 games in its league win the LCS. Seven of the seventy-two teams won at least 20 of their last 27; none won an LCS. Five of the 72 teams had losing records in their last 27 games; three of these (Braves in ’96, Florida in ’97, and Yankees in ’00) won an LCS, and two won the World Series.

Problems with the statistical analysis: (1) There are only 72 observations. (2) Correlation or, especially, linear regression, may not be appropriate for such data regardless of the number of observations, given there is more variation in the number of postseason wins than in the number of wins in the final 27 games. (3) The “samples” drawn for each team here aren’t independent. (E.g., these teams played each other in the final 27 games.) (4) Other variables may need to be included. (I did a quick google search and didn’t see any sabermetric analysis of season-end “momentum” and postseason play, but I didn’t look long.) (5) As noted, I haven’t done this type of thing in a while, so who knows what else I’m forgetting to check?

Why 27 games? Well, if I can figure out how to do this right, then it might be interesting to look at whether the timing of a team’s wins affect its post-season chances. I doubt that it does, but analyzing six segments of 27 games (6x27=162) over the full season would help pinpoint where, if anywhere, wins are more important. (E.g., how does a team whose wins are concentrated toward the beginning of the season compare with a team whose wins are more frequent at the end?) My guess—they’re all important, but there’s no reason to fret over them more at one time of the year than another.

Sleep soundly, Sox fans . . .

Johnny Mostil
09-17-2005, 09:42 AM
Here's what may be a simpler, or even better, way to consider the impact of late season "momentum" wins. Do teams with more late season wins go further into the post-season?

It doesn't seem that way. I classified the 72 post-season teams since 1996 into four categories: (1) won World Series, nine teams, (2) won LCS only, nine teams, (3) won LDS only, eighteen teams, and (4) lost in first round, thirty-six teams. The average number of late season wins (over final 27 games) for each:

World Series champs: 15.6
Won LCS only: 14.8
Won LDS only: 17.3
Lost in first round: 16.5

By contrast, "earlier" season wins matter a little more. Average number of wins in the first 135 games for each:

World Series champs: 80.3
Won LCS only: 82.2
Won LDS only: 77.9
Lost in first round: 77.6

For those of you doing the math, yes, teams that won the LCS but not the World Series won more games across the season than teams any other category. Two possible explanations for that: wild cards that have won the WS, and teams from the stronger league that year winning the WS against a team with more wins.

But, beyond the fact that late-season "momentum" doesn't seem to carry deep into the post-season, and that teams with more victories across the season appear to go deeper in the post-season (in other news, the sun rose in the east this morning), I wouldn't read much into any of this. I was just curious what the evidence was on late-season "momentum." I haven't found much so far, but welcome suggestions on how else to look at the (limited) data.

Can I change my handle to Propellerhead? Or even Tin-foil Propellerhead?

PaleHoseGeorge
09-17-2005, 10:01 AM
Nicely done. You even admit where the statistical holes in your analysis lie. Nine seasons and 72 non-independent samplings are not enough to prove your thesis, but it sure doesn't look good right now for the believers in Big Mo who have been making themselves look like complete *******s these past few days... "Yeah, but 3 times (out of 4) the team with Big Mo won the playoffs" What dopes they are... complete idiots trying to understand the numbers.

Suffice to say you have 1 million times more credibility on this matter than anyone currently sucking on Cleveland's Big Mo.

:thumbsup:

Johnny Mostil
09-17-2005, 10:41 AM
Nicely done. You even admit where the statistical holes in your analysis lie. Nine seasons and 72 non-independent samplings are not enough to prove your thesis, but it sure doesn't look good right now for the believers in Big Mo who have been making themselves look like complete *******s these past few days... "Yeah, but 3 times (out of 4) the team with Big Mo won the playoffs" What dopes they are... complete idiots trying to understand the numbers.

Suffice to say you have 1 million times more credibility on this matter than anyone currently sucking on Cleveland's Big Mo.

:thumbsup:

Thanks. Out of curiosity, I ran one final comparison: average number of late season wins by whether team won or lost first-round series. Average number for first-round losers: 16.50. Average number for first-round winners: 16.22. Ohhhhkaaaayyyy.

Like I said, I don't think any of this proves anything--but if it doesn't prove anything, then it doesn't prove anything about momentum, does it?

Or does it prove momentum is for morons? :duck:

SoxEd
09-18-2005, 07:10 PM
Now, I'm no Statistician, but it seems to me that your analysis (for which thankyou btw) shows that to find out who does well in the Post-Season, we're gona have to forget Media-led armchair predictions and watch them play the actual games.

Whodathunkit?

:D:

Johnny Mostil
09-18-2005, 07:55 PM
Now, I'm no Statistician, but it seems to me that your analysis (for which thankyou btw) shows that to find out who does well in the Post-Season, we're gona have to forget Media-led armchair predictions and watch them play the actual games.

Whodathunkit?

:D:


Indeed. Here's a theory I hope to prove next: a team that wins its last post-season game does better than a team that wins its last (n) regular-season game(s) . . .

Frater Perdurabo
09-19-2005, 11:02 AM
Indeed. Here's a theory I hope to prove next: a team that wins its last post-season game does better than a team that wins its last (n) regular-season game(s) . . .

Johnny, thank you for your analysis and refreshing honesty about the numbers.

I personally believe that the games will prove my theory that the team that scores more runs in a given game will win that game, and that the team that scores the most runs on the last day of the post-season will be crowned World Series champs!

:)

wdelaney72
09-19-2005, 11:35 AM
Thank you for the research. You certainly are more qualified in this area than I am.

One question, wouldn't we need to look at this in terms of pre-wild card and post-wild card? It would seem to me that this changes the entire landscape of the sampling.

Just a thought.

Tekijawa
09-19-2005, 11:35 AM
Can anyone draw a picture or let me Borrow their Cliff's Notes on this thread... it kinda looks interesting!

Thanks,
Teki

Johnny Mostil
09-19-2005, 11:37 AM
Johnny, thank you for your analysis and refreshing honesty about the numbers.

I personally believe that the games will prove my theory that the team that scores more runs in a given game will win that game, and that the team that scores the most runs on the last day of the post-season will be crowned World Series champs!

:)

Thanks, brother. Here's another theory that may prove momentum exists after all: a team winning a divisional series has a better chance of winning a league championship series than does the team losing a divisional series . . . and I'll bet the same thing is true of the league championship series and World Series . . .

voodoochile
09-19-2005, 11:44 AM
Indeed. Here's a theory I hope to prove next: a team that wins its last post-season game does better than a team that wins its last (n) regular-season game(s) . . .

Can you follow that up with an analysis of whether teams that score more runs in a given game do better than the team that scores less runs in that same given game?:D:


Edit: Frater beat me to it... Still, thanks for the look at the numbers. If you are looking for a new set of momentum numbers to crunch, why not take a look at how hot starts affect a teams ability to make the playoffs. I'd bet you might find some kind of correlation there (but that might just be my perception). If you could prove that early season games mean more than late season games, I would simply die laughing...

Johnny Mostil
09-19-2005, 11:45 AM
Can you follow that up with an analysis of whether teams that score more runs in a given game do better than the team that scores less runs in that same given game?:D:

For you? It's in my queue . . . :cool:

PaleHoseGeorge
09-19-2005, 11:57 AM
Can anyone draw a picture or let me Borrow their Cliff's Notes on this thread... it kinda looks interesting!

Thanks,
Teki

Here's the Cliff's Notes version.

Johnny Mostil did a regression analysis of all teams qualifying for the playoffs since the introduction of the wild-card back in 1995. Regression analysis is a very complicated (but very specific, too) statistical equation usually completed using a computer program because doing it by hand is too time-consuming, causing miscalculations. However it is highly useful for statistically proving whether outcomes have "statistical significance" or are merely the result of randomness. Anyone with even a year's worth of college level training learns all about regression analysis in statistic classes. It has application in every science, including all the hard sciences (like genetics) and applied sciences (like business administration).

What Mostil found was that there is no statistical significance to the theory that teams with lots of wins in their last 27 games do better in the first round of the playoffs. The outcomes are statistically random, suggesting Big Mo is an illusion.

He also notes that a true statistician would require more data (at least 100 trials and preferably 500) and true independent trials, too (i.e. one outcome not effecting other outcomes) like when the Sox beat Cleveland the result must be the Sox add a win while Cleveland must add a loss.

Thus the data Johnny Mostil shares is preliminary only. However anyone sucking on Cleveland's Big Mo right now is truly confused to be doing so because nothing -- so far -- indicates any good reason for them to be on their knees choking on it.

Glad to help.
:cool:

Baby Fisk
09-19-2005, 12:13 PM
I stand by my facile, "no-brainer" argument that the results of Game 1 will indicate which teams have momentum once the post-season begins. All games played before then will have become moot.

Johnny Mostil
09-19-2005, 12:20 PM
Here's the Cliff's Notes version.



Tekijawa, I probably did this bass ackwards for presentation purposes. So let's take the same numbers and present them in what may be an easier way to follow. For simplicity, I state the answers to each of the questions below more directly than the evidence really allows . . .

Q. Do teams winning more late season games, i.e., those with more "momentum," find this "momentum" carries over into the first round of the playoffs?

A. No. In fact, teams winning the LDS have won, on average, fewer games than those losing the LDS (though the difference is minimal and not statistically significant).

Q. Do teams winning more late season games find this "momentum" carries them further in the playoffs?

A. No. (EDITED) In fact, teams reaching the World Series won, on average, fewer late-season games than those not reaching it, and those winning the World Series won fewer late-season games than those losing it (though, again, differences are minimal and not statistically significant).

Q. Does regression analysis show any relation to late season wins and number of wins in post-season?

A. No. In fact, the coefficient is negative (suggesting teams winning more games late in the season are less likely to win post-season games), though, as PHG notes, there are far too few cases for meaningful analysis, the "samples" aren't independent, etc.

Let's just say there's no evidence for "momentum," meaning the null hypothesis must stand: "momentum" doesn't matter, though wins anytime still do . . .

fquaye149
09-19-2005, 02:48 PM
Now, I'm no Statistician, but it seems to me that your analysis (for which thankyou btw) shows that to find out who does well in the Post-Season, we're gona have to forget Media-led armchair predictions and watch them play the actual games.

Whodathunkit?

:D:



Are you saying talk is cheap?

:moron:
"Because I think talk is expensive"

:darkcloud:
"So do I"

Tekijawa
09-19-2005, 03:05 PM
Thanks for Dumbing it down a bit guys!

Deuce
09-19-2005, 03:40 PM
Oh... but ESPN said that they got Big Mo to and fro! Mo' Mo than you know. Guess we'll all know just how much Mo after tonights show...

:nuts:

Tekijawa
09-19-2005, 04:02 PM
Oh... but ESPN said that they got Big Mo to and fro! Mo' Mo than you know. Guess we'll all know just how much Mo after tonights show...

:nuts:

You Gonna go bro?

Sorry.

Deuce
09-19-2005, 04:18 PM
You Gonna go bro?

No. Wish I was, though. Watch Garcia throw and the Sox lead grow. Guess I'll just catch the show from FSN Ohio.

The Wimperoo
09-29-2005, 02:20 PM
Here's an article backing up pretty much what you researched. It's better to be good than have "momentum".

http://www.hardballtimes.com/main/article/ten-things-about-momentum-in-the-postseason/

brewcrew/chisox
09-29-2005, 03:16 PM
Here's an article backing up pretty much what you researched. It's better to be good than have "momentum".

http://www.hardballtimes.com/main/article/ten-things-about-momentum-in-the-postseason/


MAn I just posted a thread on this article, and didn't know you guys were talking about it here. This is what I find interesting about the Studeman article:

"From 1997 through 2000, Momentum was bad.

Each World Champions from 1997 through 2000 had a worse record in September than the season as a whole. In fact, two of those teams had September records below .500. The 1997 Marlins made it to the postseason through the Wild Card slot and became the first Wild Card team to win it all. However, you can't really say they had momentum on their side, as they played only .444 ball in September, the second-lowest September record of any World Series winner in the past 35 years.

The worst September record for a World Series winner was .433; the Yankees in 2000. In fact, the 2000 Yankees lost 15 of their last 18 games but still managed to finish first in the American League East. They then went on to beat Oakland, Seattle and the Mets for the world title.

By the way, the fourth-worst September record of any World Series champion belongs to the 2001 Diamondbacks, who played .476 ball the last month of the season. But the arms of Johnson and Schilling pulled Arizona through each postseason series until they had won it all.

The irony here is that, for all the recent focus on Momentum and the postseason, September records meant almost nothing a few years ago."

Johnny, you are basically saying the same thing. Nice job.

Johnny Mostil
09-29-2005, 03:48 PM
Johnny, you are basically saying the same thing. Nice job.

Thank you. I thought Hardball Times covered it quite nicely. For my latest (separating my shoulder patting myself on the back) on why a tight race may help more than one that's a blowout, see http://www.whitesoxinteractive.com/vbulletin/showthread.php?t=58537.