THE BOOK--Playing The Percentages In Baseball

AL Divisions looking to be fairly competitive, though unbalanced.

Fairly reasonable sans the Dodgers and Phillies haters.
vr, Xei

This is surprisingly similar to my expectations except for one data point...the phillies!!  The Pads look a little too loved.

The results indicate that the process might have worked...at the very least most of the results agree with pundit/blogger analysis.

My faith in the fans has never been stronger.

What kind of stupid computer system thinks the AL Central will be won by an 82 win team?  Oh..wait..

Seriously though, I would really like to see some adjustments for teams that had more home fans voting, or had more optimistic fans in general.  Even as is though, looks pretty good.  They don’t have the Yankees winning 40 more games than the White Sox…

Hey Tango, does Marcel make similar predictions for total team wins or is it limited to individual player projections only?

If you have each team’s individual player projections you can make team projections.  The fan ballots were exactly that.  Presumably Tango tossed everything into a model and copy/pasted the outputs.  So yes, you can make similar projections with Marcel.  You just need a nice, theoretically sound model to use.

What A Team said.  (And it was David who did it, not me.)

Even better is to just wait for SG at RLYW to run his Diamond-Mind sims.

SG posted CAIRO v.4 projections on 2/23.  I suck at posting links and don’t want to make a mess here, so I’ll just say you can head over to RLYW (currently Revenge of RLYW, due to host issues) and go to the second page.  Here’s the short short version:

AL East:

Yankees 99.4
Red Sox 93.8
Rays 93.1
Orioles 68.5
Blue Jays 64.3

AL Central:

Twins 83.2
White Sox 81.8
Indians 80.8
Tigers 80.3
Royals 67.5

AL West:

Mariners 82.6
Rangers 81.4
Angels 80.7
Athletics 77.6

NL East:

Phillies 91.7
Braves 84.5
Marlins 80.0
Mets 77.1
Nationals 73.1

NL Central:

Cardinals 90.6
Reds 85.5
Cubs 83.8
Brewers 83.8
Pirates 73.6
Astros 64.1

NL West:

Rockies 84.7
Dodgers 84.6
Diamondbacks 81.5
Giants 78.3
Padres 77.9

other than the NL East, it looks like the fan projections and CAIRO match up pretty well. 

pretty cool to see

"Hey Tango, does Marcel make similar predictions for total team wins or is it limited to individual player projections only? “

I’ve run some trials with Marcel through Diamond Mind, although depth charts will still change, but here’s how they looked as of last week.

AL East
NYA 97.8
TAM 95.8
BOS 95.2
BAL 72.4
TOR 68.9

AL Central
CHA 81.8
MIN 80.7
DET 79.2
CLE 76.9
KC 69.0

AL West
SEA 81.7
LAA 81.4
TEX 80.9
OAK 78.4


NL East
PHI 91.3
ATL 85.2
NYN 78.8
FLA 78.1
WAS 70.6

NL Central
STL 91.1
CIN 84.6
CHN 84.4
MIL 81.5
PIT 71.7
HOU 64.9

NL West
COL 85.0
LAN 84.9
ARI 81.4
SF 78.7
SD 77.7

I’ll probably kick off the full run in mid-to-late March.

Once again, it looks like the over on LAA projections is a safe bet.

I’ll say this one time.  My (extensive) experience with things that objectively look like randomness (ALA beating their projection year in and year out), yet almost NO ONE believes is random (i.e., the whole world is absolutely convinced that something is going on, e.g., Scioscia is the best manager in baseball, etc.) is that it turns out to be the former - going forward of course.

IOW, if anyone wants to take the over on all the ALA projections combined (including Vegas odds) and is willing to wager 120 to win 100, I’ll happily book it, on a friendly basis of course.

Disclaimer:  This post is not meant to break any local, state or federal laws.

The projections on the AL West’s premier franchise range range from mid 70’s to mid 80’s.  The average projection is probably 81 wins for them.  To take your offer with the 120 wager vs 100 payout they’d have to be expected to beat 81 wins about 55% of the time.  What kind of true talent projection do you need to beat 81 wins 55% of the time?  Probably an 84-85 win, I’m sure somebody can do the math and get an exact number.  So your betting position is you want the under on the highest of their team projections.  Not really any different than my position (projecting them for 83).  I’d just like the over if somebody takes the under on the 76 win PECOTA or THT projections - with a equal wager/payoff.

"I’d just like the over if somebody takes the under on the 76 win PECOTA or THT projections - with a equal wager/payoff.”

Well, of course you’ll want the over on the lowest of any team’s projections of all the systems, regardless of the team.  I’ll gladly take the under on the highest ANA (or any other team) win total of all the systems.

“What kind of true talent projection do you need to beat 81 wins 55% of the time?  Probably an 84-85 win...”

Not even close, Sean.  A team wins 1 more (or 1 less) game than its true talent around 6% of the time and it wins it’s expected mean around 6.5% of the time.  (IOW, an 81 win team wins 81 games 6.5%, 80 games 6%, and 82 games 6%.)

So an 82 win team will go over 81 wins around 53.25%, go under 40.75%, and tie (81 wins), 6%.  That is already a 4.35% profit if you lose 1.20 and win 1.00 on each wager!

I’m giving you a bargain if you think that the Angels are at least 1 win better than the average projection for all the systems…