Yankees win! Theeeee Yankees win! According to 'What If Sports'

by Ed Valentine on Oct 27, 2009 5:49 PM EDT in Analysis Comment 25 comments

Can't wait for the real games to begin Wednesday night? Well, the folks at the simulation site What If Sports have already played out the games via their highly-specialized simulation software.

And the winner? What If Sports computer models like the Yankees 72.3% of the time, most likely in six games.

I know the simulations don't mean a thing. But, they are still fun to explore. And I do like the results.

0 recs | Comment 25 comments | Add comment | Share

Story-email Email Printer Print

Comments

hmmm, simulators… can’t say I am a big fan of them. :)
The site is blocked for me at work, but I wonder what kind of simulation they ran and what kind of starting pitching matchups they used for each game. I know that AccuScore has incorrect pitching matchups on their 7 game sim. I haven’t tried to simulate out beyond Game #1 yet. But in order to win a seven game series 72.3% of the time you need to be a very heavy favorite.
vr, Xei

by Xeifrank on Oct 27, 2009 6:01 PM EDT reply actions 0 recs

Tell you what

I will play MLB The Show 2009 100 times as the Yankees against the Phillies and will tell you what the percentages I come up with. Because in all honesty, it is probably about as accurate. Now I would be interested if you guys have tested your simulators. Take the past 50 WSs and run the simulators on them and give me your predictive accuracy. I bet it would be no higher than 55 percent… or rather 5% higher than flipping a coin.

by HappyLuckyGoldenDragonNumber1! on Oct 27, 2009 8:10 PM EDT up reply actions 0 recs

I don’t think anyone is going to bother with you on this one.
vr, Xei

by Xeifrank on Oct 27, 2009 11:48 PM EDT up reply actions 0 recs

Actually I want you to bother with this

You throw these simulation numbers out with us assuming your simulator has any semblance of an accurate predictive algorithm. So why don’t you run a series of prior World Series and give us a number on the simulators accuracy. Until you prove past accuracy any prediction you give us has about as much value as rolling dice. So unless your past prediction accuracy is higher than 60%+ your simulator is essential a coin flip.

by HappyLuckyGoldenDragonNumber1! on Oct 28, 2009 8:51 AM EDT up reply actions 0 recs

Nobody has any reason to prove anything to YOU. Plus the way you have phrased your question makes no sense.
vr, Xei

by Xeifrank on Oct 28, 2009 6:39 PM EDT up reply actions 0 recs

Ok their sim picked the Phillies to win 14 to 8 in game 2?

I like making up numbers too. 75% of all the facts I spew have a 54% chance of being erroneous, of the remaining 46% there is a 12% chance of them being facetious.

by HappyLuckyGoldenDragonNumber1! on Oct 27, 2009 8:14 PM EDT reply actions 0 recs

I am 33% leaning towards thinking you are being 87% sarcastic...

by waw on Oct 28, 2009 12:27 AM EDT up reply actions 0 recs

lol good one!

Just like this new Black Panther Cologne, they say 90% of the time it works 40% of the time!

by McDaniel on Oct 28, 2009 10:42 AM EDT up reply actions 0 recs

That computer is on crack

The sample 6 games they display feature huge scoreboards. With the pitching in this, we’re going to see a lot of tight games, with a handful of runs in each game.

I always feel pressure. What I don't have is fear. -- El Duque

by LateInningRelief on Oct 27, 2009 8:28 PM EDT reply actions 0 recs

Like I said, guys

Basically meaningless stuff. Just something to pass the time until we find out what will really happen.

by Ed Valentine on Oct 27, 2009 9:20 PM EDT reply actions 0 recs

Not my thing.

Besides, I don’t want to be all excited cuz the simulaters say we’ll win and then go lose. Anything can happen at any time and you shouldn’t need a computer to tell you that.

Some people are like Slinkies. They're good for nothing, but they bring a smile to your face every time you push them down the stairs.

by TheMelkman on Oct 27, 2009 9:24 PM EDT reply actions 0 recs

Very true

by Ed Valentine on Oct 27, 2009 9:36 PM EDT up reply actions 0 recs

simulators and computers have no place in sports

wasnt it some highly intelligent computer that rated derek jeter as the worst shortstop in the league????
plus i bet two years ago this same simulator would have shown the patriots beating the giants in the sb 99% of the time

by karamjitm9090 on Oct 27, 2009 10:55 PM EDT reply actions 0 recs

"simulator and computers have no place in sports"?

Tell that to college football and the BCS.

by GMan83201 on Oct 27, 2009 11:03 PM EDT up reply actions 0 recs

A good simulator has its place. A bad one like “What If” sure does not.
vr, Xei

by Xeifrank on Oct 27, 2009 11:48 PM EDT up reply actions 0 recs

A good simulator

Please, without radar we couldn’t predict weather 12 hours from now based on simulation algorithms. You love your simulator Xei, that’s fine but until you show me you have past accuracy past 50-50% your simulator results are just noise.

by HappyLuckyGoldenDragonNumber1! on Oct 28, 2009 8:53 AM EDT up reply actions 0 recs

define “accuracy”. You seem to not understand the proper way to measure the accuracy of a simulator.
vr, Xei

by Xeifrank on Oct 28, 2009 6:40 PM EDT up reply actions 0 recs

the bcs is a whole comedy of errors

the only reason theres soo much confusion is cuz theres so many bowl games and so many stipulations. for example that whole crap about the rose bowl. if it was up to me id just rank teams by conference. then hold playoffs. each playoff game could be a bowl game. that way nobody can say they were cheated. the eventual winners wll be the bcs champions

by karamjitm9090 on Oct 27, 2009 11:29 PM EDT reply actions 0 recs

Well, these guys have allegedly picked every winner in the last year

Maybe we should accept that computer simulations can be useful?

However, there should re-run their results based on likely pitching matchups. Pedro in G2, Hamels in G3. Probably an appearance by Gaudin. And I’d like to see how they weight season performance versus more recent trends.

73% isn’t great odds for this title, but it should give a little hope.

And I think we can agree that the Patriots had at least a 90% chance of winning that Super Bowl, which came down to the Helmet Catch (the football equivalent of a grounder through Buckner’s legs).

by PortlandYankee on Oct 27, 2009 11:32 PM EDT reply actions 0 recs

"73% isn't great odds"?

Dude, being a 73% favorite is roughly equivalent to having a three times better chance of winning than your competition. I’ll take 3-1 every time. Granted, 9-1 would be better, but then again, while we’re dreaming, so would 100-1!

by waw on Oct 28, 2009 12:30 AM EDT up reply actions 0 recs