Sunday, February 13, 2011

Forecasting the Blue Jays 2011 Wins-Loss Record


The other day I made an on the fly prediction that the Blue Jays would have a losing record. This prediction was based on nothing more than a feeling I had given my expectations of production for the returning players and others replacing the players they lost. There was actually no statistical analysis done to support this hypothesis as it was just a judgement call on my part, mainly due to the lack of time I had to put into the predictions. Therefore, to be fair I spent some time today to forecast the actual number of Blue Jays wins for 2011 season using the Wins Above Replacement (WAR) princple.

For those that aren't familiar with WAR it is very worthwhile reading up on and the Fangraphs Library is a great place to start. The short version is that WAR is the number of wins a player was/is worth over the expected value of a replacement level player, and a team full of replacement level players will win approximately 48 games.

The first step in forecasting the number of wins for the 2011 Blue Jays season using WAR is to start with what we know, which is the Blue Jays won 85 games in 2010, and the following Table 1 identifies a breakdown of the  2010 team WAR values:

Table 1

Position PlayerWAR
John Buck2.9
1BLyle Overbay1.5
2BAaron Hill1.1
3BEdwin Encarnacion1.8
SSYunel Escobar/Alex Gonzalez3.6
RFJose Bautista6.9
CFVernon Wells4
LFTravis Snider/Fred Lewis2.4
DHAdam Lind-0.3
BenchMisc1.6
Starter 1Shaun Marcum3.5
Starter 2Ricky Romero4
Starter 3Brandon Morrow3.7
Starter 4Brett Cecil2.5
Starter 5Rzep/Eveland/Litsch0.6
ReliefMisc2.6
Replacement Level Wins48
Total 2010 WAR Wins90.4


So we know that according to WAR the 2010 Blue Jays should have won approximately 90 games, but in actuality they only won 85.  Well since this isn't an exact science let us say for forecasting purposes that when projecting 2011 WAR values, anything above a total of 90 will result in proportionately more than 85 wins, and anything less than 90 will result in proportionately less than 85 wins.

Now when projecting the WAR values for 2011 Blue Jays I made the following assumptions:
     1. The regular starting line-up for the 2011 Blue Jays will be as identified in Table 2 below
     2. J. P. Arencibia will have a similar WAR to that of Matt Wieters rookie season (I know this isn't an exact science but when estimating rookie values your guess is as good as mine)
     3. For Adam Lind, Aaron Hill, Yunel Escobar, Rajai Davis and Juan Rivera I am using their average WAR value over their past two major league seasons
     4. For Jose Bautista, I have proportionately reduced his value from a 52 home run hitter, to that of a 40 home run hitter (my 2011 prediction).
     5. For Travis Snider, I extrapolated his 2010 production value over 162 games
     6. For Edwin Encarnacion I have maintained his 2010 WAR value as I figured with values over the last five years of 1.7, 1.4, 1.9, 0.2, and 1.8, another season of 1.8 probably wouldn't be that far off.
     7. The Blue Jays bench, Ricky Romero, Brett Cecil and the 5th starter spot would also maintain their WAR values
     8.  Kyle Drabek's WAR projection is based on the averages of Brett Cecil's season last year (2.5) and Ricky Romero's rookie year (2.8)
     9. Brandon Morrow would have a slight increase in WAR, based purely on about 20 more innings pitched.
     10. The bullpens (which includes all other pitchers not included in the starting five) improvement for 2011 is based on the difference between the total value of Scott Downs and Kevin Gregg leaving (2.0) and the addition of Octavio Dotel, Frank Francisco, and Jon Rauch (2.2)

After plugging all of these assumptions into my spreadsheet, the resulting out comes for the WAR values of the 2011 Blue Jays are presented in the following Table 2:

Table 2

Position PlayerWAR
J. P. Arencibia1.5
1BAdam Lind1.6
2BAaron Hill2.5
3BJose Bautista5.1
SSYunel Escobar3.25
RFTravis Snider3
CFRajai Davis2.25
LFJuan Rivera1.9
DHEdwin Encarnacion1.8
BenchMisc1.6
Starter 1Ricky Romero4
Starter 2Brandon Morrow4
Starter 3Brett Cecil2.5
Starter 4Kyle Drabek2.65
Starter 5Rzep/Litsch/Other0.6
ReliefMisc2.8
Replacement Level Wins48
Total 2011 WAR Wins89.05


So there you have it, the 2011 Blue Jays will have a total WAR value of 89, which if you compare it against the 2010 value it represents approximately one less win.  Therefore based on my assumptions:

The 2011 Blue Jays Will Win 84 Games

Of course, feel free to put in any alternate values that you chose to come up with your own predictions and feel free to identify them in the comments section below.  As for my earlier prediction that the Blue Jays will have a losing record, well, eyeballing the above estimates I think the Arencibia, Romero and Escobar forecasts combined are about 4 WAR points too high, which would bring the Jays 2011 record forecast down to 80-82, which is where I was originally thinking before the statistical analysis.

1 comment:

  1. By calling it a Wins-Loss record, I assumed you were going with 161-1. As for me, I'm thinking 77-85

    ReplyDelete