Overall Team Contact Rating

[Crash Course]

See: http://www.netshrine.com/20040806.html

Discussion welcome and appreciated.

[mgoettsche]

Very interesting article. I look forward to looking at the chart when I'm on a computer with Excel. :(

Here's some thoughts/questions based on this and your other SOTS.

*The K is obviously more devastating to a team than another out, just due to the fact that a ball put in play has more opportunity for random things to happen (errors, lack of range, etc). You can't have it both ways. Folks point to K/9IP as one of the biggest indicators of a pitcher's dominance, purely because it shows that the pitcher is less reliant on his defence than other pitchers. If you buy that, then you must also buy the converse.

*That said, the K is just one component of a batter's overall production. From the "just another out" camp, folks discount the K as it relates to many players because that player does so many other things well at the plate. Thome is a great example...lots of walks (so high OBP), lots of power (so high SLG). The K is part of the package. This is in contrast to a guy like Paul Lo Duca, who is a fine hitter but hardly ever walks (low OBP) and has marginal power (low SLG).

*That said, I would take a guy like Mark Grace (marginal power, very good OBP, low K) over Jose Hernandez (better than avg power, low OBP, high K) every day of the week.

If I were to rank these players (position independent) on value to the team, it would be Thome, Grace, Lo Duca, Hernandez.

*There definitely is something to be said for being able to put the bat on the ball in the postseason. Many of the most memorable WS are determined by chance...texas leaguers, running miscues, errors. Putting the bat on the ball allows for these things to happen...K's do not. Similarly, preventing the other team from making contact limits the number of opportunities they have for a freak play to happen.


I think Beane has it right...in a short series, anything can happen. A high OBP team with a solid pitching staff will be successful over the course of the season. However, if that high OBP isn't also coupled with a low K rate, that team is highly susceptible to running into a hot pitching staff, where the difference might be an error in the field.

Good article, lots to think about...

[Crash Course]

Thanks!
And, an excellent post, IMHO, by you, above - and just as thought provoking! :D

[nyy26wc]

There is a poor correlation between hitting SO and scoring runs.

But, there is a good correlation between hitting SO and fans booing.

So, the question really comes down to whether you want to score runs or avoid booing.

[nyy26wc]

But, while there is a poor correlation between hitting SO and scoring runs, there is actually a good correlation between pitching SO and preventing them.

So, by mixing hitting SO and pitching SO, you're mixing good correlation with bad and are really coming up with nothingness.

So, what you're doing is either taking the poor correlation (hitting SO) and are using something with good correlation to prop that one up, or you're taking something with good correlation (pitching SO) and are sabatoging it by throwing in something with bad correlation.

[Crash Course]

I can't understand why a strikeout is very good for a pitcher and no big deal for a hitter.

Should not the Yin and Yang of a thing be equal?

[mgoettsche]

With regards to correlation, are you talking about an individual or as a team? As an individual, I'm sure you are right...Thome and Jose Hernandez are extreme examples and would be all over the chart. Likewise, most high K/9IP guys usually are among the league leaders in ERA. As a team though, it seems that there is somewhat of a correlation...certainly not as good as OBP to R/G, but one nonetheless.

I would think team composition generally is the kicker here, and I think there are far more Lo Duca/Grace types than there are Thome/Giambi types in the player pool. Those teams with more Grace (2001 D-Backs) types than Jose Hernandez types (2001 Brewers) will be among the leaders in OBP and R/G.

Here's some examples, showing the top and bottom quartile of K teams for a couple of years, side by side with their OBP and their R/G. (not having a spreadsheet program sucks right about now).

2000 NL (Team/K/OBP/R/G)
1. Colorado 907 .362 5.98 (1)
2. Arizona 975 .333 4.89 (10)
3. Cincy 995 .343 5.06 (5)
4. Atlanta 1010 .346 5.00 (6)
.
13. San Diego 1177 .330 4.64 (12)
14. Florida 1184 .331 4.54 (t14)
15. Milwaukee 1245 .325 4.54 (t14)
16. St Louis 1253 .356 5.48 (4)

2002 AL
1. Anaheim 805(!) .341 5.25 (4)
2. KC 921 .323 4.55 (11)
3. Boston 944 .345 5.30 (2)
4. CHW 952 .338 5.28 (3)
.
11. Min 1089 .332 4.77 (9)
12. TB 1115 .314 4.18 (12)
13. Toronto 1142 .327 5.02 (7)
14. NYY 1171 .354 5.57 (1)

Small sample size, for sure, and not a tremendous correlation, but generally speaking the lower K teams are in the upper half of OBP. Eyeballing other recent years generally supports this, particularly in the NL...the distribution is certainly more evenly spread in the AL. My guess is that there are more Giambi types in the DH role, thus evening out player type distribution for AL teams.

But we have seen several of the high OBP, high K teams flame out of the playoffs in recent years, and flame out spectacularly (Houston, StL come to mind). I think there's a case to be made that a high OBP, low K team fares a better shot in the postseason due to more balls put in play.

[Crash Course]

Agreed: http://www.netshrine.com/20040408.html

[nyy26wc]

Small sample size, for sure, and not a tremendous correlation, but generally speaking the lower K teams are in the upper half of OBP.

I disagree.

I just ran a correlation over the past 10 years between OBA and SO. There was a -.19 correlation between the two. A negative correlation between two figures indicates that, in the majority of cases, when you're good at one, you're bad at the now.

The fact that the correlation was -.19, instead of -1.00, merely shows that there are some exceptions. You just happen to have eyeballed a couple of years in which you noticed a few of those exceptions.

[nyy26wc]

I just did a correlation between SO compared to the league average and team runs scored.

The correlation came to -.01, which indicates that it's almost perfectly random whether a team is good or bad at striking out and the runs scored.

[mgoettsche]

Thanks Lee. Good stuff. Eyeballing back in the '90s definitely shows a randomness to league K leaders to R/G and OBP.

Lots to think about...

I guess my question would be: Does this change in the postseason? I would tend to think that a low K/high OBP team has a little bit of an advantage over a high K/high OBP team, mainly because of the quality of the opposition's starting pitching.

[nyy26wc]

I just ran a correlation over the past 10 years between OBA and SO. There was a -.19 correlation between the two. A negative correlation between two figures indicates that, in the majority of cases, when you're good at one, you're bad at the now.

Just noticed that stupid typo in there. It should have said bad at the other.

[nyy26wc]

I guess my question would be: Does this change in the postseason? I would tend to think that a low K/high OBP team has a little bit of an advantage over a high K/high OBP team, mainly because of the quality of the opposition's starting pitching.

I have no idea what the postseason data would show. But, with the small size sample of the series, I don't know if we can get truly good data.

I don't think what would matter is the K rates in the postseason. I think what matters there is power. A good OBA/bad power team needs to string their hits and walks together to put together a rally. The better the pitcher, the harder that is. So, I think that, in the postseason, a bit of sacrifice of OBA might actually benefit a team, provided that the decrease in OBA is compensated for an increase in SLG.

[JamesI]

I don't think what would matter is the K rates in the postseason. I think what matters there is power. A good OBA/bad power team needs to string their hits and walks together to put together a rally. The better the pitcher, the harder that is. So, I think that, in the postseason, a bit of sacrifice of OBA might actually benefit a team, provided that the decrease in OBA is compensated for an increase in SLG.
I agree here. A power team seems to have the advantage in the postseason.

[Crash Course]

Maybe it is just convential wisdom, but, if good pitching does beat good hitting, then, in the postseason, where you face better pitching, then small ball may be better than power.

Yankees out OBP and out SLG, IIRC, the Marlins in the 2003 WS, and lost.

[mgoettsche]

Lots to think about...

On the surface, it makes sense that a team's SLG would need to increase in the postseason...less hits against quality pitching, so a team needs to make the most of them. This seems to imply that teams that rely on the longball should have a better chance. However, that profile just doesn't fit many of the WS champions over the past 15 years. Then again, anything can happen in a series.

Looks like a project...I'll try to work through a few Series vs regular season stats and see if anything interesting comes out. I think at the end though, we'll see a mixed bag with hitting stats with pitching trumping all.

[nyy26wc]

When 14 out of 15 teams with the best record in baseball fail to win the World Series, the one thing we can conclude is winning correlates poorly to winning championships.

And in that kind of world, what's the point in any analysis on who will win in the postseason?

[Crash Course]

I dunno. There must be a way to see what kinds of teams typically do well in the postseason and then apply that to future fields, no?

[mgoettsche]

I think you're pretty close. Pitching and high OBP are a good formula for getting to the postseason. Once a team makes it though, its pretty random as to what will happen. Look at the past three series:

NY - .261/.332/.406; 2.13 ERA; 5 E
Fla - .232/.281/.300; 3.21 ERA; 2 E

NY outhit (outslugged, out-OBP'd) and outpitched Fla by over a run per game. Florida struck out one more hitter and struck out one less time at the plate...pretty even there. For all of Fla's vaunted speed, both teams stole exactly the same number of bases. From an ink perspective, errors appear to have been decisive.

Ana - .310/.369/.465; 5.75 ERA; 5 E
SF - .281/.362/.498; 5.55 ERA; 5 E

Anaheim had 10 more hits, but this appears to be offset by SF 7 (!) more HRs. SF struck out 12 more times than the Angels (50 to 38). Pretty even with regard to OPS and ERA...perhaps the Ks were the difference?

Ari - .264/.312/.394; 1.94 ERA; 3 E
NY - .183/.237/.288; 4.26 ERA; 8 E

Wow, did this series really go seven games? Arizona had a decisive edge at the plate...23 more hits, same number of HRs. You would think NY would have struck out more than the D-Backs, but this wasn't the case...Arizona struck out 70 times, Yanks only 63 (still pretty close).

The only uniform thing about these three series is that the team that committed the fewest number of errors won.

[Sidewinder]

Ari - .264/.312/.394; 1.94 ERA; 3 E
NY - .183/.237/.288; 4.26 ERA; 8 E

Wow, did this series really go seven games? Arizona had a decisive edge at the plate...23 more hits, same number of HRs. You would think NY would have struck out more than the D-Backs, but this wasn't the case...Arizona struck out 70 times, Yanks only 63 (still pretty close).


We have Byung-Hyun Kim to thank for that.