Playoff experience/Winning

by scarymonsters85 on Nov 15, 2008 11:56 AM CST

David Gassko has an interesting article in the HBT: http://www.hardballtimes.com/main/article/does-experience-matter-in-the-postseason/

about whether or not having playoff experience helps you win in the playoffs. He concludes that it does.

I posted this to get some chatter out of the usual suspects( cwyers, dakoose et al).

I personally feel that the study was flawed because generally good teams will make the playoffs more often than poor teams. A team that makes the playoffs one year, will most likely repeat that success in the following years. Take the yankees for example. A team that is consistently good for 10 plus years, wins a couple of championships; they will rack up play off experience, as opposed to the marlins (x2), the dbacks, the mets and a couple of other teams they competed against. It would be ill advised to conclude that the reason they won these series' and had playoff success is due to having playoff experience.

For me to actually accept the study, there would have to be nearly overwhelming evidence to support it. It is really hard to predict an outcome of a short series ,or a 7 game series, REALLY hard. Playoff experience, as a metric for superiority is highly suspect in my book.

What do you all think?

This is a FanPost and does not necessarily reflect the views of SB Nation or Al Yellon, managing editor (unless it's a FanPost posted by Al). FanPost opinions are valued expressions of opinion by passionate and knowledgeable baseball fans.

Tweet Comment 3 comments | 0 recs |

Do you like this story?

Comments

I feel the same way about that article as you do.

Also, I would also be interested to know what exactly is so beneficial about playoff experience, not just whether it makes a difference or not? Perhaps, like you said, teams that make it more often than others do better because they are better, and therefore have a better chance at succeeding against the higher quality opponents that the playoffs are stocked with. If experience means being surprised by the level of intensity, shouldn’t just a game, maybe two, be enough to get over it? Gasko’s article is definitely interesting, but I didn’t walk away thinking any differently than I did before reading it. The benefit of experience may be something that can never be proved or disproved.

by dakoose on Nov 15, 2008 1:16 PM CST reply actions

There's no way to measure the underlying pathway...

through which experience would affect probability of winning. You can only measure whether or not there exists some pathway (which is what this analysis is trying to do).

For the record, even if the “shock” hypothesis only holds for one or two games, that’s a HUGE deal in a seven-game series. It would be interesting to explore the probability of winning early games as a function of playoff experience. Again – it won’t definitively prove anything, but it could present strong support (or strong evidence to the contrary) of that hypothesis.

But in terms of determining whether or not playoff experience has any sort of relationship, you run regressions like this. The key is controlling for other factors that might be correlated with probability of winning (especially if they may also be correlated with experience). The big one would be talent, which I would hope would be well proxied by record. Unless you can come up with another variable that significantly associated with probability of winning (and assuming the regression was run properly), then I think the results are defendable.

by SouthernCub on Nov 15, 2008 1:45 PM CST up reply actions

I think they control for exactly what you are talking about...

It appears that the author did in fact control for the team’s talent level to some degree. He includes Pythagorean record in the regression, which should be controlling for any difference in talent that you might expect to see from a team that has more playoff experience.

If that is the case, then the article is presenting exactly what it suggests: that playoff experience has a positive effect on probability of a series win even after controlling for talent difference.

I’d need to see the actual output of the regressions run to be more certain, but if he did consider Pythagorean record then that should address some of your concerns. I’d have probably also run the analysis on actual record just in case there was some omitted variable that causes actual record to be different than Pythagorean record.

by SouthernCub on Nov 15, 2008 1:35 PM CST reply actions