Anybody out there familiar with "Context Adjusted Defense"

[misterdirt]

I moved this comment to its correct spot at the end of the thread.

[misterdirt]

I moved this comment to its correct spot at the end of the thread.

[terpsfan101]

I recently came across an intriguing article titled "Context Adjusted Defense" on Baseball Think Factory. It was written a few years ago by Charles Saeger. This seems like it would be a very solid Defensive metric, but nobody has really mentioned it since it was published a little over 4 years ago. Charles gives you all the formulas that are necessary for calculating the adjustments, however, the presentation of the methods is beyond confusing. I tried to set up a spreadsheet to follow along with him, but get lost when he gets to the part about position rates and adjusting opportunities for left-handed and right-handed batters. Has anybody else attempted to go through these calculations? Or maybe somebody could help me figure out the adjustment for right/left handed batters faced. Here is the link to the article for those who are interested in learning more.

http://stats.mostvaluablenetwork.com/the-goodthe-bad-and-the-context-adjusted/

[misterdirt]

Reading Mitchel's (MGL's) comments at the end of the article might help to sort things out for you. But then again, MGL admits that he is somewhat confused about the details himself. Although CAD seems to be a useful attempt to estimate defensive ability for the years for which we have no PBP data, it makes no sense to use it for any year that we do have PBP data. PBP data provides exact numbers for opportunity (or at least more accurate estimates) so defensive metrics that utilize PBP data will always be more accurate.

[SABR Matt]

I don't like that he arbitrarily throws out half of the defensive equation for infielders. Putouts are outs too...it may be true that infield putouts require much less skill than infield assists, but they still need to be accounted for...and it is possible to make a fairly accurate assessment of how many putouts are unassisted and should therefore count as plays made. And oh BTW, it's erroneous to claim that there's no skill in catching infield flies. Robby Alomar and Derek Jeter are both examples of players who demonstrate skill in that area...they might suck at everything else, but they can range for fly balls and that does impact the scoreboard.

[terpsfan101]

I think CAD would still be useful for those of us who do not how to parse the play by play data from Retrosheet. After I run the BGEDIT program on the data, and import it into MS Access, I have no idea how to go about compiling it. I believe the play by play data tells you the hit location of the ball, whether it was a grounder, line drive, etc..., the runners on bases before the play, runners removed after the play, and finally the actual result of the play (hit, error, out). CAD tries to estimate opportunities for fielders in 4 areas: Range (Assists for IF/Opportunities and PO for OF Opps), Runners Removed/Available (DP's and DP Opps for IF and Assists for OF), Error Rates, and finally IF putouts. So Misterdirt is right, a PBP database could give us an exact number for all these opportunities.

[terpsfan101]

Referring to SABR Matt's post:

I know Derek Jeter takes a lot of heat for his defense, and justifiably so. These are just my personal observations of his defense.

The Good:

-He is the best SS I have ever seen when it comes to making a relay throw to gun down the baserunner.
-He does an excellent job of fielding ground balls when he has to take a few steps in.
-As SABR Matt pointed out, he has a lot of range when it comes to catching popouts and flyouts.
-He is OK at turning the DP.
-He has an OK arm

The Bad:

-He has absolutely no range when he has to move laterally to field a ground ball. A lot of basehits go through the hole at short and up the middle when you watch Yankee games.
-His positioning in the field before the play is terrible also. He never seems to be in the right place before the play.

This is my two cents on Jeter's defense. Overall, I'd say he's average to slightly below average in the field. The positives I mentioned above, pretty much negate his terrible range when he has to move laterally and his poor positioning.

[SABR Matt]

terps...in the last couple of seasons, the statistics back you up. Jeter has improved at double play turning and making plays to prevent baserunner advances enough that it's starting to eat into the negatives he has (the total lack of range from side to side and the poor positioning) and turn him into a near-average fielder. When he was young though, he hadn't developed those skills enough and when you total up his whole career, he still ranks as one of the worst defensive shortstops of all time. Perhaps that will change if he can stay average for several more years. Thing is...he's starting to lose his natural speed so his range is getting worse.

[terpsfan101]

Alright, if someone out there could help me work through the RHB/LHB adjustment for context-adjusted defense, I would be grateful.

OK, lets look at the AL shortsops on the team level for the 1998 season.

The first thing Charlie attempts to measure is range using this formula:

Player outs / (Player outs + team hits allowed)

For shortstops player outs is simply assists made.

ANA: 454
BAL: 520
BOS: 466
CHI: 488
CLE: 494
DET: 538
KC: 549
MIN: 456
NYA: 440
OAK: 495
SEA: 453
TB: 525
TEX: 483
TOR: 447
LgAverage: 486

Charles now tries to estimate the number of singles that infielders are responsible for. To do this we need the teams total number of singles allowed, and there groundball ratio. Let's calculate the groundball ratio first.
Here is the formula for that:

Groundouts: A - A.c - A.of - DP.1b
Flyouts: PO - SO - A

and this is the formula for the ratio:

GO/(GO+FO).

So for each team this looks like:

Tm, GB, FB, GO/(GO+FO) ratio
ANA, 1404, 1596, .468
BAL, 1517, 1474, .507188231
BOS, 1393, 1656, .456871105
CHI, 1454, 1696, .461587302
CLE, 1475, 1616, .477191847
DET, 1533, 1578, .492767599
KC, 1439, 1619, .470568999
MIN, 1353, 1808, .428029105
NYA, 1395, 1649, .458278581
OAK, 1414, 1728, .450031827
SEA, 1340, 1547, .464149636
TB, 1457, 1556, .483571192
TEX, 1344, 1720, .438642298
TOR, 1302, 1709, .43241448
LgAVG, 1416, 1639, .463502455

Now that we have our GB ratio, we apply this to the total number of singles allowed, to estimate how many singles the infield was responsible for allowing:

ANA: 967 1Ballow * .468 = 453
BAL: 1033 1Ballow * .507188231 = 524
BOS: 951 1Ballow * .456871105 = 434
CHI: 1024 1Ballow * .4615873 = 473
CLE: 1049 1Ballow * .47719185 = 501
DET: 1046 1Ballow * .4927676 = 515
KC: 1079 1Ballow * .470569 = 508
MIN: 1082 1Ballow * .4280291 = 463
NYA: 921 1Ballow * .45827858 = 422
OAK: 1025 1Ballow * .45003183 = 461
SEA: 949 1Ballow * .46414964 = 441
TB: 967 1Ballow * .48357119 = 468
TEX: 1083 1Ballow * .4386423 = 475
TOR: 907 1Ballow * .43241448 = 392
LgAVG: 1006 1Ballow * .463502455 = 466

Ok, so here is what we know so far:

TM, Outs, Hits IF Responsible for
ANA: 454, 453
BAL: 520, 524
BOS: 466, 434
CHA: 488, 473
CLE: 494, 501
DET: 538, 515
KC: 549, 508
MIN: 456, 463
NYA: 440, 422
OAK: 495, 461
SEA: 453, 441
TB: 525, 468
TEX: 483, 475
TOR: 447, 392
LgAVG: 486, 466

Now, before we figure the range figure we need to make the adjustment for RHB and LHB faced. And here is where I get lost. I know the adjustment gets applied to failed opportunities (in this case, singles allowed), but in his formula, I don't know what he means by "league average." We are interested in the second version of the adjustment. Here is what he says about the adjustments:

"With all this in mind, I present two versions of the adjustment. The first is for seasons in which we do not have LHB/RHB data:

((BFP.rhp - HR.rhp - HB.rhp - BB.rhp - SO.rhp) / (BFP.tm - HR.tm - HB.tm - BB.tom - SO.tm)) / lgAVG) ^ 0.25

This is for third basemen. Divide by 2 and add 0.50 for shortstops, and divide by 5 and add 0.80 for right fielders. For first basemen, divide the third base rate into one (or, more mathematically, this is the inverse); for second basemen, divide the shortstop rate into one; and for left fielders, divide the right field rate into one.

The second is for seasons for which we do have LHB/RHB data:

(((AB.rhb - HR.rhb - SO.rhb) * posRate) / (AB.tm - HR.tm - SO.tm)) / lgAVG

What the heck is the posRate? It is the “position rate,” a measure of how much more likely a right-handed batter is to hit a ball to that position. Each affected position has a different rate:

1B 0.25
2B 0.50
3B 4.00
SS 2.00
LF 0.70
RF 1.30

A right-handed batter is four times as likely to hit a ball to the third baseman than a left-handed batter, and twice as likely to hit a ball to the shortstop than a left-handed batter. This sounds like a big difference, and it is, but teams don’t vary much on the number of right-handed batters they face.
On the flip side (pun intended), that same batter is 30% more likely to hit the ball to the right fielder. That’s not a huge difference, and I debated on whether to adjust for that. I chose to do so, since it’s not entirely trivial. Do not make this adjustment when you are making estimates of left/center/right playing time."

Like I said earlier, I don't know what he means by LgAvg. If you use this as LgAvg: (LgABrhb - LgHRrhb - LgKrhb) / (ABlg - HRlg -Klg), the results always come out the same, no matter what you plug in for the position rate part of the equation.

Anyway here is the data you need to do the adjustment. Maybe somebody could help me figure out how it is calculated. To save space, I already solved for the (AB-HR-K) part of the equation, and will list that as Balls in Play. So we have:

TM, BIPrhb, BIPtotal
ANA, 2502, 4293
BAL, 2283, 4301
BOS, 2282, 4318
CHI, 2707, 4529
CLE, 2224, 4461
DET, 2435, 4464
KC, 2676, 4458
MIN, 2531, 4583
NYA, 2437, 4248
OAK, 2493, 4527
SEA, 2683, 4244
TB, 2413, 4287
TEX, 2481, 4537
TOR, 2093, 4309
LgAVG, 2446, 4397

Again, I would appreciate somebody's assistance in figuring out this adjustment.

[misterdirt]

terpsfan101 - Derek Jeter: -He is the best SS I have ever seen when it comes to making a relay throw to gun down the baserunner.

Its interesting how our perceptions are not always supported by the facts. I ran a query on the 2003-2005 retrosheet fielding data. For those three years there were 389 assists by shortstops on balls fielded by outfielders, an average of about 13 per team. During that time Jeter had only 7 assists. 16 shortstops had more (as many as 22) and 3 others also had 7. Jeter is not only not the best at the relay throw, he is not even in the top half. Now some of this may be a lack of opportunity, i.e. the runners respect his arm enough that they don't try to take the extra base on him. But I doubt it since the main factors on taking an extra base are the hit location, the speed of the runner, the quality of the arm of the fielder fielding the ball, and the game situation. The quality of the arm of the relay man is not much of a factor in limiting opportunities.

[terpsfan101]

Wow, this is the main point I always used to argue that Jeter wasn't that bad of a defensive shortstop. I came to this conclusion during Game 7 of the 2001 World Series, when he gunned down an Arizona Diamondback at third base on an incredibly acrobatic relay throw. This was also the year that he made that amazing throw/flip to nail Jason Giambi at home plate in Game 3 of the Division series. Alright, I'll stop before it sounds like I'm a Jeter fan. Still, watching Jeter make a relay throw is a thing of beauty. But you could be right, just because it looks pretty, doesn't necessarily mean it's more effective.

[misterdirt]

Maybe Jeter was making up in the post season for his regular season performance. Here's Jeter's line for 2001 regular season.

150 games
1321.1 innings
343 assists
0 assists on relay throws from the outfield

Yes, you are reading that correctly. He didn't have a single assist on a relay throw from the outfield during the entire regular season. When I did the query I thought I had done something wrong when I found 126 outfield relay assists for all the shortstops but Jeter's name didn't show up at all. So I went back and checked each assist that Jeter recorded for the season. All 343 assists that MLB had credited for Jeter were in retrosheet's PBP data and not one was a relay throw from an outfielder.

[SABR Matt]

Maybe Jeter was making up in the post season for his regular season performance. Here's Jeter's line for 2001 regular season.

150 games
1321.1 innings
343 assists
0 assists on relay throws from the outfield

Yes, you are reading that correctly. He didn't have a single assist on a relay throw from the outfield during the entire regular season. When I did the query I thought I had done something wrong when I found 126 outfield relay assists for all the shortstops but Jeter's name didn't show up at all. So I went back and checked each assist that Jeter recorded for the season. All 343 assists that MLB had credited for Jeter were in retrosheet's PBP data and not one was a relay throw from an outfielder.

LOL! Now that's funny. It's amazing how a player can acquire a reputation for doing things he NEVER does just by making a couple of spectaculra plays in the post-season.

[Gubanich Plague]

Still, watching Jeter make a relay throw is a thing of beauty. But you could be right, just because it looks pretty, doesn't necessarily mean it's more effective.

I know exactly what you mean. You know how as an infielder, when you're receiving a throw from an outfielder, you're supposed to do a little crow hop so that you're shifting your weight towards the infield as you make the relay throw? Well, when Jeter does that, he lifts his legs REALLY high on his crow hop, and it makes the whole thing look, well, quite elegant.

I think that's his style. Everything's just exaggerated with him. The jump throw from the hole at shortstop, the way he jackknifes away from a pitch inside, and don't even get me started on that popup he caught against the Red Sox and then dove into the seats.

Now I actually do think he's got great range in catching popups, and I think he might make a pretty good first baseman.

[SABR Matt]

I think Jeter would be an outstanding center fielder.

[Mariano_Rivera]

I think Jeter would be an outstanding center fielder.

Do you think it might have been a better idea for the Yankees to play Jeter in CF, A-Rod at SS, signing a 3B FA and allowing Eric DUncan to continue on his career path than forcing him to move to 1B :rolleyes:

Everybody moves for "Captain Clutch" even if it's worse for the team.

[Mariano_Rivera]

I know exactly what you mean. You know how as an infielder, when you're receiving a throw from an outfielder, you're supposed to do a little crow hop so that you're shifting your weight towards the infield as you make the relay throw? Well, when Jeter does that, he lifts his legs REALLY high on his crow hop, and it makes the whole thing look, well, quite elegant.

I think that's his style. Everything's just exaggerated with him. The jump throw from the hole at shortstop, the way he jackknifes away from a pitch inside, and don't even get me started on that popup he caught against the Red Sox and then dove into the seats.
Now I actually do think he's got great range in catching popups, and I think he might make a pretty good first baseman.

LOL no kidding when I first heard about that play i evisioned him diving into the seats to catch ball otherwise impossible to catch. Instead it was more like him catching a ball in LF and then the montum took him into the seats. Somehow every time I watch that catch I think why didn`t he slide into the wall feet-first rather than cutting up his face

[SABR Matt]

Jeter makes everything look a little more spectacular than it actually is...you have to give him credit for being a showman...it provides hours of entertainment for Yankee fans everywhere.

In terms of winning games? Not so much...

And yes, when the Yankees acquired A-Rod I advocated sliding Jeter to CF, keeping A-Rod at short, DHing Bernie Williams and finding a free agent third baseman.

[misterdirt]

Now I actually do think he's got great range in catching popups, and I think he might make a pretty good first baseman.

This is a case where your impression is correct. The yankees get a lot of pop ups hit to the right side and Jeter makes the play on almost all of them. For 2003-2005 there were 283 balls pop ups or short fly balls hit to shortstop, in short left field and behind 3d base including the foul territory behind 3d. Jeter got all but 6 of them. The bad news is that even though this is a real skill and one that Jeter is actually good at, there isn't much difference between regular shortstops as to getting to and catching popups. Angel Berroa had the most popups fall in for hits or errors of all the regular shortstops from 2003-2005 and he only had 13. The difference between Berroa and Jeter is only 2 plays not made a year. Most shortstops fall within + or - 1 play a year which is less than + or - 1 run a year. This amount of difference is just as likely to be the result of chance as skill. And certainly not really enough to earn you a gold glove if you can't get to the ground balls hit to your left or right.

As for Jeter's signature play of catching popups in foul territory behind third, he was indeed among the best. He made 12 catches in foul territory which was only topped by Berroa and Guzman with 13 and Lugo with 14.

I'm not sure about Jeter as a centerfielder. Being a good centerfielder requires both raw speed and quick reactions. He definitely has (or had) the raw speed, but one has to wonder about his reactions given his inability to get to the ground balls hit to his right or left. And pasta diving Jeter in center would be an inside the park home run. It would have been an interesting experiment though, and probably no worse for the team than keeping him at short.

[Tango Tiger]

Jeter's skillset is rather unique, if we base it on how Fans evaluate him:
http://www.tangotiger.net/scouting/sim2006_5406.html

There is no one really like him, and only two that are somewhat like him (the equally-maligned SS Michael Young and 2B Luis Castillo). The rest of his comps are so-so comps, and are really all over the place, in LF, RF, 3B.

You can reasonably argue just about any position for him, though the corner positions (3B, LF, RF) probably make the most sense to anyone who is not a coach of MLB.

Jeter will not be a SS in his mid-30s, but will still be an excellent hitter. Just a matter of time before he moves somewhere. At Yankee Stadium, it's easier to play RF than LF (much more ground to cover in LF than RF as the peak distance is deeper in the LF/CF gap than RF/CF), so he'll probably end up in right. His strong arm will help.

[Gubanich Plague]

Pasta diving?

[Chisox]

Pasta diving?
I think it's supposed to read "And past a diving Jeter." Forgot a space, it happens. For a couple seconds though, I was wondering the same thing.

[Gubanich Plague]

Wow, there we go. Yeah, for the life of me I couldn't figure that one out.