Handicapping MLB Win Totals

slogger · #1 03-07-2005, 02:09 PM

I was wondering if anybody had some insight on what data to consider when projecting total regular season wins for MLB teams.

Trends such as taking unders for popular teams (NYY, BOS, CHC, ATL)? Importance of hitting vs. pitching? Importance of avg. age of players?

I'm looking to develop a system whereby I am able to project a certain win total for each team and then use that projection to determine whether any available Win Totals are worth betting.

slogger · #2 03-08-2005, 02:00 AM

*bump*

craig r · #3 03-08-2005, 02:35 AM

I have no idea how to handicap them, but from what i can gather, the value is generally in the under. I think if you add them all up, they usually add up to more games than there are in a season. Sorry, I couldn't be more help.

craig

Clarkmeister · #4 03-08-2005, 02:51 AM

I have a hard time believing I can't do more with my money in a 6 month timeframe than betting on a MLB total with maybe a 10-20% edge. Surely you can find many such spots in other events between now and October? Seems like a waste to me unless you are really overbankrolled.

Jeff W · #5 03-08-2005, 06:01 AM

Runs allowed and Runs earned are the only stats you need to project win totals. Here is an article from Baseball Prospectus.

The next step is projecting runs allowed and earned player by player for a given team. Add the totals and use the forumla from the article to project a W-L record. Compare the projected W-L total to the W-L total for the book and then calculate your edge to determine if you can be the vig.

critterdude · #6 03-08-2005, 10:07 AM

[ QUOTE ]
I have a hard time believing I can't do more with my money in a 6 month timeframe than betting on a MLB total with maybe a 10-20% edge. Surely you can find many such spots in other events between now and October? Seems like a waste to me unless you are really overbankrolled.

[/ QUOTE ]

I agree but there is another exception other than over bank rolled.

With contract style betting where you can sell anytime (preferably for a nice profit) you can get great value within a good turn around.

You do need to handicap for a base of reference and have the information of the timing so that you buy just before a jump in price.

This type of betting does require a little more skill and time. The advantage is softer lines perhaps because the smart money is betting in the more traditional method rather than the stock market variation.

DougOzzzz · #7 03-08-2005, 11:29 AM

[ QUOTE ]
Runs allowed and Runs earned are the only stats you need to project win totals.

[/ QUOTE ]
Great! Please direct me to the 2005 runs allowed and runs scored statistics for MLB teams.

[ QUOTE ]

Here is an article from Baseball Prospectus.

The next step is projecting runs allowed and earned player by player for a given team. Add the totals and use the forumla from the article to project a W-L record. Compare the projected W-L total to the W-L total for the book and then calculate your edge to determine if you can be the vig.

[/ QUOTE ]

Easier said than done. Even a sophisticated projection system like PECOTA on BP seems a little out of whack when projecting runs scored/runs allowed. For instance, the past couple years the Yanks and Sox have been projected to win ~105 games.

I think the problem with this is that more things can go wrong to good teams than bad... or at least if something does go wrong it has a greater effect on their expected wins.

Paluka · #8 03-08-2005, 01:54 PM

[ QUOTE ]
I think the problem with this is that more things can go wrong to good teams than bad... or at least if something does go wrong it has a greater effect on their expected wins.

[/ QUOTE ]

I don't think this is true at all. Tigers were pretty bad 2 years ago, but boy o boy did things get worse. Same goes for Diamondbacks last year.