How to estimate playing time

by
March 31, 2012

The real key to forecasting performance in any year is getting a player’s playing time right. Sure, averages and percentages rise and fall, but they remain outside our best efforts to make them more predictable. Playing time is tough to predict too, but something can be done here. This is where you come in.

People who follow teams have real insight into who is most likely to play where. So Tangotiger has started his annual Playing Time Survey to get your input. Just head on over to his site, pick your favorite team, and enter your best guess as to who is going to play how often during the season. This is a community project; the results are available to all. So please help by contributing. It will just take a couple of minutes.


Dave Studeman was called a "national treasure" by Rob Neyer. Seriously. Follow his sporadic tweets @dastudes.
4 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Steve I
12 years ago

Hey, Dave,
There’s a theory in science-fiction that polling very large numbers of people results in remarkably accurate estimates.  Some questions (How much did a WW II battleship cost?) draw absurb answers ($50,000), but the final result is likely to be close to true.
I don’t endorse the theory, but it seems plausible.
Questions:
Have you heard of this theory?  Or am I just out of my skull?  (I can’t cite a source, but I think I’ve read it in a couple places, maybe called The Delphi Theory.)
Do experts do better at prognostication than semi-random, very large (on the order of millions), samples?  (I would expect experts to do better than random people on questions of the past (or present), but maybe not better on the future.  E.g., an expert would know a battleship cost $4,000,000 in 1935, but might be way low or high on predicting what it would cost to build one in 2035.)

0
Dave Studeman
12 years ago

Hey Steve,

Yes, that’s exactly the theory behind what Tango is doing here.  I know it as “The Wisdom of the Crowd.”  Here’s a wikipedia link on the subject:

http://en.wikipedia.org/wiki/Wisdom_of_the_crowd

0
napster
12 years ago

The wisdom of the crowd works fine when there are a sizeable number of people who are knowledgeable of whatever questions are being answered.  But when there are a lot of ignorant people involved, or when there are very few knowledgeable persons this so called wisdom falls apart.

The “crowd” can also be motivated by hysteria or lack of reliable information.  Unforeseen events can also occur which instantly skews the so called conventional wisdom.

Since those who estimate the playing time are mostly knowledgeable persons, this method will work pretty well.

0
JFree
12 years ago

napster

Knowledge of those who vote/poll is precisely why a poll like this is going to produce pretty good results compared to “experts”

Actual voting for politicians by voters who are completely ignorant of the actual issues is why we elect politicians who are stupid/venal/corrupt and why parties/experts can so easily manipulate us

0