Category Archives: Usenet

Mirror of posts to alt.sports.baseball.chicago-cubs

NLCS and Nate Silver

On Friday, October 21, 2016 at 1:23:11 AM UTC-5, Frank Sereno wrote:

> Nate Silver’s Fivethirtyeight.com has installed the Cubs as
> favorites to win the World Series after last night’s win at 51%
> while the Indians are at 39 and the Dodgers at 10.

That’s interesting.  The Vegas lines on this would be far more accurate than Nate Silver.  I don’t have them however.

Although Nate Silver’s  election analysis is the best in industry because polling data can predict the future, he struggles with sports and much of it is simply made up.  It would be interesting to see all his math on this.  Some of the writers who work for him have published complete nonsense with their sports analysis but I digress…

The next two games for the Cubs should be around 50/50.  Kershaw should be favored in Game 6 like he was in game 2 but Arrieta should be equally favored the next game.  To keep the math simple let’s just say even steven both games.

The Dodgers have a 25% chance of going to the World Series, Cubs 3/4 or 75%.

If Nate is giving the Cubs 50% to win the WS that means his handicap for them is  66.6% favorites for the series against Cleveland.  This is quite high.  He thinks Dodgers would be almost even with Cleveland.

For LAN:  0.25 * 0.40 =~ 10%
For CHN:  0.75 * 0.67 =~ 50%

P(CLE) = P(Opponent = LAN)*( 1 – P(LAN)) + P(Opponent = CHN) * (1 – P(CHN))

where P(TeamID) is probability that team wins the series after winning the NLCS.

We know P(LAN) = 0.40 , P(CHN) = 0.67
P(Opponent = LAN) = 0.25
P(Opponent = CHN) = 0.75

Plugging all those numbers in I get Cleveland should be 40%.

> I have to work Saturday so at least I won’t have to hear Joe Buck
> verbally fellate Clayton Kershaw.

It will be funny to listen if Clayton has a bad game which is very much in the realm of possibility.  I will miss Larry King however.Watching Larry squirm and check his watch was priceless.  I hope I never have to see his face on TV  again.

Step away from the ledge

NOTE:  Ironically when I started making this data model in 2013 I covered the St. Louis Cardinals and Boston Red Sox extensively.  Now that my Cubs have a legit shot at a World Series I haven’t done much here at all.  This is because the data model is finished and although data comes in every day I still haven’t found a way to beat the lines put out by Vegas or ESPN.  Now that the Cubbies are down 2-1 I made this analysis.

The Cubs just have to win one in LA to get back home field advantage.  Step away from the ledge.  Here is today’s line…

DATE 10_19 8:05_PM Oct_19_10:02 CHN LAN
LINEAWAY CHN [ 0.545 ] < 0.512 >
STARTAWAY 3.46 John_Lackey_CHN
LINEHOME LAN [ 0.500 ] < 0.512 >
STARTHOME 1.39 Julio_Urias_LAN

At WAA=1.39 Urias is unranked, out of the top 200 so he’s average.  According to an analysis I did awhile ago Cubs do well against average pitchers.  Rich Hill and Kershaw are well above
average.  The Cubs had problems with way above average and for some reason  way below average pitchers (but there won’t be any of them).

The line started giving the Cubs home field advantage but now it’s in we absolutely have
no idea territory.  Flip a coin.  Heads the Cubs win, tails they lose.  It’s that close.

If both games are even steven the probability for LAN to win the next two games are:

P(win next two) = (1/2)*(1/2) = 1/4 or 25%

Same for the Cubs.  So the probability of Cubs returning to Wrigley this year is 1-0.25=0.75 or 75%.  Not too complicated. This series is far from over.  Whoever wins today changes these probabilities drastically.

Here was the Kershaw game line ….

DATE 10_16 8:05_PM Oct_16_18:02 LAN CHN
LINEAWAY LAN [ 0.567 ] < 0.554 >
STARTAWAY 8.57 Clayton_Kershaw_LAN
LINEHOME CHN [ 0.452 ] < 0.467 >
STARTHOME 8.88 Kyle_Hendricks_CHN

The Ouija board had the Dodgers as clear favorites. They didn’t care that Kershaw wasn’t Clayton Kershaw in the playoffs.  I think that when Dave Roberts put Kershaw in to close out the NLDS in a real live game situation that did something to Kershaw’s psyche.  That’s a trait which can   earn one a Manager of the Year trophy.  Let’s hope that’s the only trophy LAN gets.

If the Cubs make it back to Wrigley hopefully Kershaw reverts back to playoff Kershaw.  Then
the Cubs should be OK.

Jay Bruce to Mets

On Monday, August 1, 2016 at 11:40:36 AM UTC-5, The old geezer wrote:
> Jay Bruce to the Mutts!

First things first.  Who TF is Jay Bruce?

2008 0.8 Jay_Bruce_CIN RF-CF-LF XXXXX
2009 1.4 Jay_Bruce_CIN RF +223+
2010 2.5 Jay_Bruce_CIN RF +126+
2011 4.1 Jay_Bruce_CIN RF +056+
2012 4.9 Jay_Bruce_CIN RF +029+
2013 5.5 Jay_Bruce_CIN RF +015+
2014 2.1 Jay_Bruce_CIN RF +149+
2015 1.9 Jay_Bruce_CIN RF +164+
2016 4.9 Jay_Bruce_CIN RF +009+
Total = 28.1

He’s a very very good player and this
year he’s #9 in MLB.  He’s at Bryant/Rizzo
level right now.  Before you start with
the sky is falling, the Mets batting
completely sucks except when they play
the Cubs this season.  Let’s look at the
Mets shall we?

BAT|PITCH|Rs|Ra|W|L|UR|LR|TeamID
-77.4 86.8 381 372 54 50 5.4 -5.8 NYN

As you can see at -77 BAT, that sucks.
Their real WAA=W-L= +4 so they’re barely
treading water but it looks like they
still want to make a run for it.  With
that PITCHing they probably can if they
improve hitting.  Bruce will be a huge help.

The Mets did this last season when they
acquired Cespedes.  Here’s a recent lineup.

GAME COL NYN July_30_2016 7:10_PM
———————————-
-2.23 Curtis_Granderson_NYN 413 RF
-0.84 Neil_Walker_NYN 370 2B
2.16 Yoenis_Cespedes_NYN 367 LF
-0.10 Wilmer_Flores_NYN 217 3B
-2.16 Asdrubal_Cabrera_NYN 398 SS
-1.39 Travis_d’Arnaud_NYN 141 C
0.00 Justin_Ruggiano NO_ENTRY CF
-1.99 Kelly_Johnson_TOT 210 1B
-0.55 Bartolo_Colon_NYN 39 P
TOTAL= -6.93

Granderson is having a very bad year and he’s leading
off for some reason.  By replacing Granderson with
Bruce -2.25 goes off the books and +4.9 WAA goes on
the books.  Combined Mets will add +7.15 value to their
lineup bringing it to even steven.

The Mets could get into the playoffs through the back
door with this move.

1938 Cubs and Gabby Hartnett

On Monday, June 13, 2016 at 4:10:29 PM UTC-5, The old geezer wrote:
> ....hit the "Homer in the Gloamin'" back in '38!!!
>
> TOG
>
> ND: Southern Tier 2X IPA
> NP: She Can't Find Her Keys - Paul Peterson

OK.  You made me lookup homer in the gloamin and Wikipedia
has an entire page on it.

https://en.wikipedia.org/wiki/Homer_in_the_Gloamin'

Here are top Cubs today back in 1938.

BAT|PITCH|Rs|Ra|W|L|UR|LR|TeamID 06/14/1938
50.6 20.3 285 201 31 21 15.6 0.9 CHN

BAT a little better than PITCH.  UR=15.6 so
they had very good fielding.  W-L = 31-21=+10
--------------------------------
+006+ 3.78 Bill_Lee_CHN PITCH
+016+ 2.77 Augie_Galan_CHN LF-OF
+035+ 2.04 Joe_Marty_CHN OF-CF
+043+ 1.81 Gabby_Hartnett_CHN CR
+044+ 1.76 Tony_Lazzeri_CHN SS
+046+ 1.72 Dizzy_Dean_CHN PITCH

The above would be top 100 today.  I had to
look this up but manager Grimm gets fired
mid season and replaced by Gabby Hartnett.
Here's how they finished the season to make
it into the World Series.

BAT|PITCH|Rs|Ra|W|L|UR|LR|TeamID
-32.2 121.1 713 598 89 63 28.8 -2.9 CHN
--------------------------------
+004+ 10.12 Bill_Lee_CHN PITCH
+017+ 6.66 Clay_Bryant_CHN PITCH
+038+ 4.85 Charlie_Root_CHN PITCH
+047+ 4.07 Dizzy_Dean_CHN PITCH

BAT tanks and PITCH takes off.  They were
lucky to win a pennant with only 89 wins.
Where's Gabby?

+098+ 1.39 Gabby_Hartnett_CHN CR

It must have been difficult for Gabby to
both manage a team and hit because from
now until the end of the season he declines
around 0.5 WAA and loses rank in the process.

The walk off homer helped get them in
the WS but they weren't as good as
these guys.

BAT|PITCH|Rs|Ra|W|L|UR|LR|TeamID
107.8 -72.9 862 795 84 70 25.8 6.1 DET
--------------------------------
+002+ 13.33 Hank_Greenberg_DET 1B
+008+ 8.25 Rudy_York_DET CR-LF-OF
+010+ 7.92 Charlie_Gehringer_DET 2B

There are three guys ending the season
in the top ten. 

Update:  Status lines between DET and CHN in the 1938 World Series show two diametrically opposite teams.  DET has tremendous BAT, CHN below average BAT.  CHN has tremendous PITCH and DET is way below average in that category.  The old saying that defense wins championships wasn’t true in 1938.  It was offense that beat the Cubs.

As we saw in last year’s playoffs, team status lines can deceive.  New York Mets posted an almost completely average BAT  for the year yet their lineup was way way above average because they made deals throughout the season to improve.   It is possible in 1938 DET improved their starting pitching significantly through trades.   If they just improved it to average that would have been significant.

It is probable that CHN having such a high BAT before AS break or trade deadline or however they did things back then caused them not to improve anything with their hitting so that below average -32 BAT probably reflected their playoff roster.

I could drill deeper in this but 1938 is water under the bridge.  Later in the season this data model keeps  track of lineup WAA, relief WAA, and bench WAA.   WAA has additive properties so the value of a set of players is simply the sum of their WAAs.  For each team we can know the WAA for starter, relievers, lineup, and bench for each team.  How these 4 categories interact with the opposition categories to determine a winning probability is still a work in progress.

Dexter Fowler

On Friday, June 10, 2016 at 11:25:44 PM UTC-5, Mordecai Brown wrote:
> I take a snapshot of current year data every
> other day.  This Dexter Fowler WAR fascinated
> me because he did hit #10 in my model earlier
> so I have been watching his
> progress closely all year.
>
> I had to modify some code to make this
> report and now I can do it for any player
> any year.  Here is a table of partial lines
> for Fowler.  I wanted to show BA and OBP
> because they too confirm  my observation
> about Fowler.  I don't compute
> OPS and it would be too much work for
> now to recalculate rank because I forgot
> to save that data off and have to write
> a script to recreate it.  
>
> Here's the daily data.  Date is the snapshot
> date which includes all data up to the
> date before.  So 0414 includes all games up
> to 4/13 of this year.
>
>
> Date|WAA|BA|OBP|NAME_TEAMID|POS
> ------------------------------------------
> 0414 0.57 0.455 0.586 Dexter_Fowler_CHN BAT
> 0416 0.53 0.423 0.559 Dexter_Fowler_CHN BAT
> 0418 0.55 0.424 0.558 Dexter_Fowler_CHN CF
> 0420 0.78 0.375 0.510 Dexter_Fowler_CHN CF
> 0422 0.82 0.375 0.508 Dexter_Fowler_CHN CF
> 0424 1.18 0.393 0.521 Dexter_Fowler_CHN CF
> 0426 1.47 0.385 0.506 Dexter_Fowler_CHN CF
> 0428 1.36 0.385 0.506 Dexter_Fowler_CHN CF
> 0430 1.26 0.368 0.488 Dexter_Fowler_CHN CF
> 0502 1.43 0.347 0.474 Dexter_Fowler_CHN CF
> 0504 1.34 0.350 0.470 Dexter_Fowler_CHN CF
> 0506 1.41 0.352 0.473 Dexter_Fowler_CHN CF
> 0507 1.26 0.340 0.462 Dexter_Fowler_CHN CF
> 0509 1.26 0.337 0.455 Dexter_Fowler_CHN CF
> 0511 1.39 0.340 0.462 Dexter_Fowler_CHN CF
> 0513 1.28 0.339 0.467 Dexter_Fowler_CHN CF
> 0515 1.20 0.333 0.458 Dexter_Fowler_CHN CF
> 0517 1.28 0.331 0.447 Dexter_Fowler_CHN CF
> 0519 1.20 0.331 0.449 Dexter_Fowler_CHN CF
> 0521 0.97 0.319 0.443 Dexter_Fowler_CHN CF
> 0522 1.09 0.315 0.441 Dexter_Fowler_CHN CF
> 0524 1.22 0.324 0.445 Dexter_Fowler_CHN CF
> 0526 1.05 0.314 0.435 Dexter_Fowler_CHN CF
> 0528 1.22 0.319 0.436 Dexter_Fowler_CHN CF
> 0529 1.07 0.318 0.432 Dexter_Fowler_CHN CF
> 0531 1.13 0.316 0.429 Dexter_Fowler_CHN CF
> 0602 1.11 0.316 0.433 Dexter_Fowler_CHN CF
> 0604 0.97 0.313 0.435 Dexter_Fowler_CHN CF
> 0606 0.76 0.305 0.427 Dexter_Fowler_CHN CF
> 0608 0.76 0.303 0.421 Dexter_Fowler_CHN CF
> 0610 0.90 0.307 0.422 Dexter_Fowler_CHN CF
>
> He's now at 0.90 WAA down from his high
> of 1.47 on 4/25.  His BA/OBP was also
> 0.393 /0.521 then and it's down to 0.307/0.422.
> Even old school stats confirm my model.
> Why WAR has him ranked so high is a mystery.
> He's almost in the top ten in WAR ranking
> right now.
>
> It would have been interesting to show
> his daily WAR but I don't have that data
> and can't even reproduce it as its generation
> is also a mystery.
>
> And finally, at 0.9 Fowler is doing well.
> My issue is he's not top of the league
> right now and maybe shouldn't even
> be in the AS game this season.  He
> very much helped make the Cubs so dominant
> in April with his bat.  I'm not dissing
> on Fowler here at all.

I want to show Rizzo who has a WAR=1.8 when
Fowler's WAR was 3.0.  Rizzo has had a very
low BA but a decent OBP throughout the year.
He ends up yesterday at +2.25 WAA compared
to Fowler's 0.9.  Who is right about this?

Date|WAA|BA|OBP|NAME_TEAMID|POS
======================================
0414 1.01 0.240 0.424 Anthony_Rizzo_CHN
0416 0.97 0.207 0.395 Anthony_Rizzo_CHN
0418 0.80 0.189 0.348 Anthony_Rizzo_CHN
0420 0.82 0.186 0.352 Anthony_Rizzo_CHN
0422 0.69 0.163 0.339 Anthony_Rizzo_CHN
0424 1.36 0.186 0.333 Anthony_Rizzo_CHN
0426 1.36 0.185 0.346 Anthony_Rizzo_CHN
0428 1.78 0.203 0.360 Anthony_Rizzo_CHN
0430 1.68 0.205 0.356 Anthony_Rizzo_CHN
0502 1.97 0.218 0.384 Anthony_Rizzo_CHN
0504 1.78 0.220 0.385 Anthony_Rizzo_CHN
0506 2.16 0.242 0.395 Anthony_Rizzo_CHN
0507 2.37 0.253 0.398 Anthony_Rizzo_CHN
0509 2.60 0.257 0.409 Anthony_Rizzo_CHN
0511 2.60 0.270 0.413 Anthony_Rizzo_CHN
0513 2.54 0.281 0.423 Anthony_Rizzo_CHN
0515 2.46 0.275 0.424 Anthony_Rizzo_CHN
0517 2.88 0.279 0.419 Anthony_Rizzo_CHN
0519 2.79 0.273 0.409 Anthony_Rizzo_CHN
0521 2.65 0.259 0.395 Anthony_Rizzo_CHN
0522 2.46 0.253 0.392 Anthony_Rizzo_CHN
0524 2.37 0.245 0.382 Anthony_Rizzo_CHN
0526 2.31 0.239 0.374 Anthony_Rizzo_CHN
0528 2.10 0.234 0.371 Anthony_Rizzo_CHN
0529 2.04 0.234 0.371 Anthony_Rizzo_CHN
0531 1.89 0.229 0.368 Anthony_Rizzo_CHN
0602 2.02 0.239 0.378 Anthony_Rizzo_CHN
0604 1.81 0.238 0.377 Anthony_Rizzo_CHN
0606 2.04 0.246 0.385 Anthony_Rizzo_CHN
0608 2.21 0.251 0.390 Anthony_Rizzo_CHN
0610 2.25 0.260 0.392 Anthony_Rizzo_CHN