Let's Catch Those Thieves

[Lane]

Inspired by Ron's recent feature on steals, I thought I'd dig in and look at the stealing from the defensive side. Who's running on who, and who's throwing them out.

The big picture: Stolen Base Attempts Against by team

SBAAbT.png (15.55 KiB) Viewed 1003 times

Whoa, holy Hawaii, Batman! *opens OOTP, navigates to HAW, hovers over Paul Scholes*

Oh, OK. So, teams don't like to run against a guy with an 11 arm. Makes sense.

In fact, if you look at the four teams with the lowest stolen base attempts against (SBAA), you'll find the only four teams that carry catchers with arms rated as "11"; HAW, LV, CAL, LOU.


So what does the Caught Stealing % look like?

CSbT.png (14.61 KiB) Viewed 1003 times

Hmmm...no real correlation here with the previous chart. HAW, LV look good, but LOU and CAL are downright bad! DM, PHX and JAX stand out though. In DM, Cisco Arreola is having a career year, throwing out 41% of runners, and his mate Jose Moreno is throwing out 44% despite just a 7 rated arm. Of the catchers on the best CS% teams, only Dave Robertson of JAX has a 10 arm, with the others in the 7-8 range.

I'm not really sure what to say about the lack of correlation between the two charts, other than maybe a smaller sample size for the teams with the least amount of attempts. Or perhaps only the best runners dare run on the best throwing catchers, tipping the scales for that reason.

Now, to borrow a bit from Ron's method, how many runs are gained or lost via the defense against the steal?

RGLbT.png (11.5 KiB) Viewed 1003 times

Keep in mind that for this analysis a CS is worth +.64 runs and a SB is worth -.33 runs. Of course this lines up well with the previous chart, with PHX and DM excellent CS% leading to them gaining the most runs via controlling the running game (over two wins each). Surprisingly, despite excellent catcher arms, and few attempts, CAL is near the bottom, losing over a half a win due to the "ineffectiveness" of their defense of the steal.

The bottom line here is (of course) in line with Ron's conclusion that overall, teams are losing runs due to steal attempts. This shows the inverse, that overall, teams are gaining runs due to poor base stealing. Charm City, in particular should be thrilled to be leading the league in SBAA, since it's helped them add over two wins to their season by this measure.

[Ted]

I think the interpretation for CAL and LOU is what you said, that only the best base stealers run on us, therefore the CS percentage is artificially low. The real question is, are you better with a catcher with a slightly worse arm, who entices the bad runners to steal and then get thrown out. Hmmm. Also, I'm not sure you can go with a CS being worth +.64 runs and a SB being -.33 runs. Can you just invert those from their offensive value to the run prevention side?

[Lane]

Addendum

What's the sum total of teams stealing on offense and defense you ask? Well I crunched the numbers.

TOD.png (13.22 KiB) Viewed 996 times

Phoenix looking great, lots of teams doing well.

Yo, Halifax. Get it together.

[Lane]

Also, I'm not sure you can go with a CS being worth +.64 runs and a SB being -.33 runs. Can you just invert those from their offensive value to the run prevention side?

I'm assuming it's just an inversion of the offensive contributions. No idea if that's correct though.

[RonCo]

That inversion should be right enough.

Interesting data.

- A few odd thoughts. There will be more "offensive" SB than defensive SB Allowed due to double steals (I think when a double steal happens, each runner gets credited with a steal, but the catcher only gets on SB allowed). You can see this in the data.
- Not surprising...the "defensive" value of running game prevention is mostly positive (because most teams are losing runs by running so often). For example, my own YS9 catchers are "positive" in run prevention, but we are below average relative to the league.

[ae37jr]

Brooklyn is going to keep running till we steal the Landis

....might take a few years though.

[RonCo]

Other thoughts:

- Assuming those weights are reasonable, except for a couple outliers, the impact of the running game is no more than a game or two either way.
- Our teams are running too often, so even poor catchers are throwing folks out enough to "make" runs.
- Part of this is probably that we think an "8" should be good, but it's not.
- It's my proposition that a 10/9 Speed/Steal combination is the worst of all worlds. I should do something to check that, I suppose.
- Another thing not being discussed is that a pitcher hold rating is important in keeping SB rates down, too. I've looked at it before. It does make a difference, though how big is hard to say. This is hard to study.

[Ted]

I guess the reason I was wondering about the inversion is this: If more teams are losing runs by running to much, running prevention is bad, and if more teams are gaining runs by running too much, running prevention is good. It seems like any real analysis of the cost of a steal or CS has to take that into effect.

Also, the schedule influence on this makes this sort of analysis tough. The FLP is easily the best running division in the Brewster, with only one team really hurting themselves on the base-paths, one neutral, and five well over. Every other division has multiple teams that are really hurting themselves. Notice that the top NINE teams in runs gained via CS are in the Johnson. That's almost assuredly due to the schedule rather than catcher arm strength.

[Ted]

Other thoughts:

- Assuming those weights are reasonable, except for a couple outliers, the impact of the running game is no more than a game or two either way.
- Our teams are running too often, so even poor catchers are throwing folks out enough to "make" runs.
- Part of this is probably that we think an "8" should be good, but it's not.
- It's my proposition that a 10/9 Speed/Steal combination is the worst of all worlds. I should do something to check that, I suppose.
- Another thing not being discussed is that a pitcher hold rating is important in keeping SB rates down, too. I've looked at it before. It does make a difference, though how big is hard to say. This is hard to study.

Right, there are just so many factors that influence each other that makes it hard to isolate the effectiveness on one player. Since largely the guys with good hold ratings are lefties and bad ones are righties, it would be interesting to how CS percentage an runs rates differ by number of lefties in the starting rotation.

[RonCo]

The values do change if you weight SB/CS differently. The .33/.64 weighting results in a cross-over point of just short of 66%.

To the HR question, the value changes in ballparks, too. In max parks, where runs are harder to come by, the value of a steal could be much higher (as could be the cost)...in a park where homers happen a ton, the cost could be higher and the value lower. (who cares if you steal second and third if the guy homers behind you?). Every GM gets to weigh these himself.

[udlb58]

So what does the Caught Stealing % look like?

CSbT.png

Hmmm...no real correlation here with the previous chart. HAW, LV look good, but LOU and CAL are downright bad! DM, PHX and JAX stand out though. In DM, Cisco Arreola is having a career year, throwing out 41% of runners, and his mate Jose Moreno is throwing out 44% despite just a 7 rated arm. Of the catchers on the best CS% teams, only Dave Robertson of JAX has a 10 arm, with the others in the 7-8 range.

I'm not really sure what to say about the lack of correlation between the two charts, other than maybe a smaller sample size for the teams with the least amount of attempts. Or perhaps only the best runners dare run on the best throwing catchers, tipping the scales for that reason.

I'd be interested to see what the average hold ratings are of those teams at the bottom of the stolen bases against list. Maybe that is why Jacksonville is higher in the CS%? (could just be dumb luck and a good season)

Robertson's arm has quietly been very important to the success of the Hurricanes. I know the league environment was different back then, but he allowed just 19 steals against in 100 games in 2029 and he has been a big deterrent to steals against the past two seasons. Still, I was finding it hard to shell out $6 million+ over 4+ years to keep that deterrent behind the plate. My recent trade means I can allow him to test free agency and look at keeping him if the market is right.

Brooklyn is going to keep running till we steal the Landis

....might take a few years though.

Come at me bro. 8+ average HLD among my top 4 SPs, 10+ arm from my two catchers

[RonCo]

A simple data pull from the stats page in-game:

Code: Select all

Team	#Pit	Hold (Avg)
DM	13	9.08
SEA	14	8.93
MTL	14	8.14
ATC	12	7.92
PHX	13	7.85
NO	14	7.71
LV	14	7.43
CAR	14	7.36
CLG	13	7.31
TWC	13	7.23
BRK	15	7.20
HNT	14	7.14
LBC	17	7.12
JAX	14	7.07
OMA	14	7.07
YS9	14	7.07
HFX	16	6.94
SA	13	6.85
VAN	17	6.76
EDM	14	6.71
LOU	14	6.64
MAD	16	6.44
CAL	12	6.42
VAL	17	6.29
CCJ	14	6.29
IND	14	6.29
HAW	15	6.27
HAV	17	6.06

[RonCo]

There is very little doubt in my mind that pitcher hold affects both SB Attempts and SB success rate to some degree.

[usnspecialist]

A simple data pull from the stats page in-game:

Code: Select all

Team	#Pit	Hold (Avg)
DM	13	9.08
SEA	14	8.93
MTL	14	8.14
ATC	12	7.92
PHX	13	7.85
NO	14	7.71
LV	14	7.43
CAR	14	7.36
CLG	13	7.31
TWC	13	7.23
BRK	15	7.20
HNT	14	7.14
LBC	17	7.12
JAX	14	7.07
OMA	14	7.07
YS9	14	7.07
HFX	16	6.94
SA	13	6.85
VAN	17	6.76
EDM	14	6.71
LOU	14	6.64
MAD	16	6.44
CAL	12	6.42
VAL	17	6.29
CCJ	14	6.29
IND	14	6.29
HAW	15	6.27
HAV	17	6.06

o look, yet another category my pitchers suck at... I guess I should be happy with gamboa that he is middle of the pack in throwing guys out despite his pitchers not doing much to help him out.