Sabermetrics | Introduction and Discussion

[Boxcar]

Thanks for clarifying. One more if you don't mind. I've been looked but haven't gotten clear on it.

 

Statistically evaluating it, how bad was Jeter at SS on D over his career?

 

One of the worst SS defenders to ever play the game. Him staying at SS when A-Rod came over was laughable.

 

Exactly how bad he was is hard to say because DRS/UZR were not around for a large part of his career (and while he was young.)

[Abomination]

Thanks for clarifying. One more if you don't mind. I've been looked but haven't gotten clear on it.

 

Statistically evaluating it, how bad was Jeter at SS on D over his career?

 

I think he was something like -3 wins, but early stats weren't very comprehensive so that's mainly hand waving. Keep in mind that up until now, we haven't had enough data to really determine individual defense fairly (granted Jeter has a large enough body of work that it is probably pretty accurate for his career). With the new data systems now in place, we have defender's position, ball trajectory and velocity, speed, defender reaction times, etc (well we may not, but teams do). It should be possible to determine exact defensive ability to very good accuracy going forward.

[Nox]

You should look into the Julia language as well as R. (http://julialang.org/) I haven't had time to develop my stuff yet (my other project is almost finished though I think, so I should be able to soon). I'm planning on trying out Julia when I do my pitchfx database and analysis programs.

 

Julia is cool but it's raw as hell. The tooling is complete garbage too.

 

It's almost C equivalent speed wise though so it will be interesting if it gains momentum.

[GD]

Julia is cool but it's raw as hell. The tooling is complete garbage too.

 

It's almost C equivalent speed wise though so it will be interesting if it gains momentum.

 

Post count of 2000? Probably just a troll

[Abomination]

Some courses on R, and using it with stats and machine learning. Looks like the intro course and a few other select ones are free. For the rest you have to subscribe ($25/month, cancel any time). I haven't gone through any yet so I can't speak to the quality, but thought people might be interested.

 

https://www.datacamp.com/courses/free-introduction-to-r

[TilsonBritoFan]

Here's one reason we have the best run differential in the league and best offense in terms of runs scored, but are sitting below .500. Lets compare some Jays/Yankees stats.

 

# of games we scored 6+ runs: 42 out of 91 games. We score 6 or more runs 46% of the time! The Jays went 35-7 in those games

# of games NYY scored 6+ runs: 27 out of 88 games. NYY score 6+ runs 31% of the time. Yankees went 23-4 in those games

 

# of games we scored 2 or less runs: 23 out of 91 games. Due to poor pitching this year we almost never win when scoring 2 or less. The Jays went 2-21

# of games NYY scored 2 or less runs: 25 out of 88 games. NYY managed to go 2-24

 

# of games we scored between 3 and 5 runs: 26 times. The Jays record in these games is 9-17.

# of games NYY scored between 3 and 5 runs 36 times and went 23-13 in those 35 games.

 

To summarize:

 

The Yankees, like most teams score between 3 and 5 runs in the majority of their games. They have a slightly better pitching staff than the Jays and they managed to win 64% of those games.

The Jays score between 3 and 5 runs less often, and when they do they have a terrible 35% win percentage. This is essentially the difference between these 2 teams. This can't all be pitching...is it...luck?

 

Lets look at the other teams in the East when they score between 3-5 runs per game:

TAM: 25-15 .625

BOS: 17-18 .486

BAL: 19-12 .612

NYY: 23-13 .639

TOR: 9-17 .346

 

As you can see our Jays truly are getting destroyed compared to other teams in the division when they score between 3 and 5 runs per game.

 

Lets take a look at each teams winning percentages in games where they don't score 3-5 runs per game:

TOR: 36-29 .537

NYY: 25-27 .481

BOS: 25-29 .463

BAL: 25-32 .439

TAM: 21-30 .420

 

The Jays would be on top.

 

These games with 3-5 runs are the games they need to win. Almost everyone loses 90% of the games where they score 2 or less. Almost everyone wins 85% of the games where they score 6+. The Jays are comparable there.

 

They are getting killed when they score between 3 and 5 runs.

Edited July 13, 2015 by TilsonBritoFan

[Dick_Pole]

So...what you're saying is the Jays need to score 6+ a game 100% of the time and they'll have a winning percentage similar to the Golden State Warriors at home! Time to trade Navarro and demote Goins.

[Spanky99]

This was posted not long ago by Twistedlogic, just saying. I'd go with a combination of luck SP and D...our pen is in the top 5 in xFIP, SP bottom 5.

 

Edit: Holy crap, our Pytho is at 54-37! Definitely luck, SP and D, lol. *sighs*

Edited July 13, 2015 by Spanky99

[TilsonBritoFan]

Sorry didn't know Twistedlogic had posted something similar already

[Boxcar]

Sorry didn't know Twistedlogic had posted something similar already

 

Too late. Cops are en route.

 

Seriously though, someone has to take Navarro away from Gibbons. I am struggling to search for his angle. Navarro sucks at everything and it's not like he has a shiny BABIP like Colabello or anything that's giving a mirage boost to his numbers. I don't think managers matter much, as I've said, but a big way they can matter is by benching good players in favour of s***** ones. Justin Smoak has deserved to play every day.

[Governator]

Is there a stat somewhere that shows the average # of pitches thrown / IP for a pitcher? Not that it really matters but I'm curious how the better pitchers who go deep in games work their pitch count. It's also useful to know how much abuse young pitchers have actually taken by pitches thrown compared to just IP or what relievers might actually be available based on how much work they did their last game (5 pitches for 3 outs is a lot different than 30).

[Abomination]

Is there a stat somewhere that shows the average # of pitches thrown / IP for a pitcher? Not that it really matters but I'm curious how the better pitchers who go deep in games work their pitch count. It's also useful to know how much abuse young pitchers have actually taken by pitches thrown compared to just IP or what relievers might actually be available based on how much work they did their last game (5 pitches for 3 outs is a lot different than 30).

 

ESPN.com has it under their Expanded II Pitching Stats. You could also just download a spreadsheet of the players from Fangraphs and calculate it in Excel.

[Abomination]

Looks like the NHL has partnered with MLBAM to provide stats, streaming video, and web services for hockey. It'll be interesting to see what data we'll get access to.

[WillyWonka]

Anyone have any idea how valid Baseball Reference's WAR calculations are? The Jays' 2nd MVP by WAR is Kevin Pillar with 3.2....a healthy lead over Russell Martin at 2.6. Obviously, most of this comes from 2.1 defensive WAR vs 1.3 offensive....which bothers me because oWAR + dWAR does not equal WAR. Defensive stats are always a question mark (Joeybats with a dWar of -0.9???), but can this actually be true?

[TwistedLogic]

FanGraphs WAR is less blasphemous.

 

http://www.fangraphs.com/leaders.aspx?pos=all&stats=bat&lg=all&qual=0&type=8&season=2015&month=33&season1=2015&ind=0&team=14&rost=0&age=0&filter=&players=

[Deadpool]

I usually use fWAR over bWAR, personally... fWAR has Pillar as the 5th most valuable Jay, which is still pretty goddamn awesome...

 

1 Josh Donaldson 5.9

2 Russell Martin 3.0

3 Jose Bautista 2.5

4 Devon Travis 2.4

5 Kevin Pillar 2.1

 

That's still pretty amazing for a guy we expected to be a 4th OF as best going forward.

[John_Havok]

Explaining the differnce in how bbref calculates WAR versus fangraphs would take a while, but the short answer for me is, any way if calculating value that has Kevin Pillar as the second most valuable Jays is probably flawed.

 

That being said, he is having a very good defensive year, and Bautista is not even close to being this great defender the talking heads on TV try to convince people of evry game.

[WillyWonka]

lol @ Tulo already being #8 according to both after 9 games.

[GD]

lol @ Tulo already being #8 according to both after 9 games.

 

15 WAR pace, nbd.

[WillyWonka]

That being said, he is having a very good defensive year, and Bautista is not even close to being this great defender the talking heads on TV try to convince people of evry game.

 

But that arm....

 

He's not great, but an entire loss attributed to his defence?

[John_Havok]

His arm is really his only positive defensive attribute. He doesn't take great routes, his range is geting smaller by the month and even his arm is hurting this year after that ridiculous attempt to throw out that runner at first.

 

I've never thought of Bautista as an overal good defender. His reputation as a plus defender has never been backed up by the eye test or defensive metrics.

[WillyWonka]

I've always seen neutral range with a + arm. Still...can stats factor in running attempts that were NOT made out of fear of the howitzer?

[Atothe]

His arm is really his only positive defensive attribute. He doesn't take great routes, his range is geting smaller by the month and even his arm is hurting this year after that ridiculous attempt to throw out that runner at first.

 

I've never thought of Bautista as an overal good defender. His reputation as a plus defender has never been backed up by the eye test or defensive metrics.

 

He has good jumps, good routes, decent speed but a average arm

[baloojayz]

I usually use fWAR over bWAR, personally... fWAR has Pillar as the 5th most valuable Jay, which is still pretty goddamn awesome...

 

1 Josh Donaldson 5.9

2 Russell Martin 3.0

3 Jose Bautista 2.5

4 Devon Travis 2.4

5 Kevin Pillar 2.1

 

That's still pretty amazing for a guy we expected to be a 4th OF as best going forward.

 

How is Bautista below average defensively? Has a cannon of an arm, decent speed?

[Atothe]

How is Bautista below average defensively? Has a cannon of an arm, decent speed?

 

terrible range and below average speed

[baubau]

terrible range and below average speed

 

Pretty much. Now with his shoulder injury this year his arm may not be a factor in the near future.

 

The dude can still throw and I've seen enough games where the 3B coach stop a player just out of fear of his arm. Teams do not challenge him often.

[Spanky99]

Anyone have any idea how valid Baseball Reference's WAR calculations are? The Jays' 2nd MVP by WAR is Kevin Pillar with 3.2....a healthy lead over Russell Martin at 2.6. Obviously, most of this comes from 2.1 defensive WAR vs 1.3 offensive....which bothers me because oWAR + dWAR does not equal WAR. Defensive stats are always a question mark (Joeybats with a dWar of -0.9???), but can this actually be true?

 

bWAR is garbage

[dineke]

I usually use fWAR over bWAR, personally... fWAR has Pillar as the 5th most valuable Jay, which is still pretty goddamn awesome...

 

1 Josh Donaldson 5.9

2 Russell Martin 3.0

3 Jose Bautista 2.5

4 Devon Travis 2.4

5 Kevin Pillar 2.1

 

That's still pretty amazing for a guy we expected to be a 4th OF as best going forward.

 

Travis da gawd

[Jimcanuck]

without the howitzer Jose goes from a average (arm compensating for his lack of range) to a below average defender. he's probably bottom 5 in among MLB rightfielders with the shredded arm.

 

and runners are taking advantage, seeing it again and again

 

Jose is quite weak at tracking balls over his head (he's fine with balls hit in front of him) and he has a fear of the wall, you rarely see him go up against it and a fair number of long balls that another outfielder will run down fall for doubles or triples

 

Pompey in RF and Jose at 1B with EE as DH is the best team alignment at the moment IMO

[Angrioter]

Anyone have any idea how valid Baseball Reference's WAR calculations are? The Jays' 2nd MVP by WAR is Kevin Pillar with 3.2....a healthy lead over Russell Martin at 2.6. Obviously, most of this comes from 2.1 defensive WAR vs 1.3 offensive....which bothers me because oWAR + dWAR does not equal WAR. Defensive stats are always a question mark (Joeybats with a dWar of -0.9???), but can this actually be true?

 

BREF puts far too much weight on defense; Brendan Ryan is like Barry Bonds for them.