Observing Clutch

[visick]

Cummings...

I'm also using Catalanotto in a league starting tonight @ PNC.

LMK how he does for you.

visick


PS-He's got "sick" clutch. :lol: (in a good way...)

[pedakrla]

[quote:902681ee5d="socalchiro"]IMHO partly true...

A single or double with a man on 2nd. and 2 outs in Petco is the same as a single (or double) with a man on 2nd. and 2 outs in Coors...No?[/quote:902681ee5d]

Yes.....and no. Let's say over an entire season, poor clutch performance loses you 25 runs in Petco, and 25 runs in Coors. Those 25 runs you lose in Petco have a larger negative impact on your W-L record than the 25 runs you lose in Coors.

[LANCEBOUSLEY]

aray,

i see your problem with so... positive clutch is not the only thing to consider when choosing a number 5 hitter. total clean xbh is another item especially if you have managed to put some speed in the top two spots or have someone like AROD batting 3rd.

another thing is how many clean hits the guy has sans clutch. there are an equal number of opportunities to drive in a run from second with less than two outs are there are with two outs. so if you have someone who has 15 hits with out clutch and 20 with i'd just as soon have the guy who has 20 w/o and 15 wtih clutch who has 10-12 clean xbh,(should they exist)

[packleader]

For me clutch is best used for what it prompts HAL to do, ie, regardless of the quality of the hitters, HAL seems to pitch to a non-clutch batter rather than a clutch batter in a 2 out situation with runners in scoring position. What this means is that even if you put a mediocre clutch hitter behind say, a Victor Martinez who is batting 5th, HAL will pitch to Martinez rather than walk him to face the clutch batter hitting 6th. Likewise, HAL will walk the mediocre clutch hitter to face the better hitter without clutch. I will take my chances in these two scenarios since then my better hitter gets to hit instead of being walked. Having said all that, I like to put two or three clutch hitters back to back to back in the 5, 6, and 7 slot in the lineup if possible as I believe it helps to create big innings which steals wins that you occasionally don't really deserve.

[Mean Dean]

[quote:0b6bd5d17a]I like to put two or three clutch hitters back to back to back in the 5, 6, and 7 slot in the lineup if possible as I believe it helps to create big innings which steals wins that you occasionally don't really deserve.[/quote:0b6bd5d17a]
I'm not sure there is much reason to think that the 6-8 spots get any unusually high number of clutch opps. The results of my study indicated that the frequency is 5th, 9th, 4th, 8th, 6th, 7th (and, honestly, it showed less dramatic differences between [i:0b6bd5d17a]any[/i:0b6bd5d17a] of the slots than may have been expected.) Now, admittedly there could be other reasons for that; maybe, for instance, if HAL hit the best clutch guy 5th more often, then 5th would be making fewer 3rd outs and there'd then be more clutch situations available to 6th and 7th. So I'm not saying that's gospel, but I dunno if there is evidence to the contrary either.

[quote:0b6bd5d17a] I can see batting Jeff Kent fifth, but I have a hard time doing so with So Taguchi, for example. Is it me, or would Taguchi be better off batting seventh or ninth? Or should I just suspend disbelief and bat him fifth whenever I've got him? [/quote:0b6bd5d17a]
This is another way of phrasing the "how many runs is it worth" question, IMO. How much better does Generally Good Hitter (GGH) have to be than Not Quite As Good In General But Excellent In The Clutch Hitter (not gonna bother with the acronym ;) in order for GGH to be the better #5 hitter? If we could estimate how many runs the clutch rating is worth, then we could give a pretty precise answer to that, since we can easily estimate how many runs both players create on the year setting clutch effects aside.

[quote:0b6bd5d17a]For me clutch is best used for what it prompts HAL to do, ie, regardless of the quality of the hitters, HAL seems to pitch to a non-clutch batter rather than a clutch batter in a 2 out situation with runners in scoring position. What this means is that even if you put a mediocre clutch hitter behind say, a Victor Martinez who is batting 5th, HAL will pitch to Martinez rather than walk him to face the clutch batter hitting 6th. Likewise, HAL will walk the mediocre clutch hitter to face the better hitter without clutch. I will take my chances in these two scenarios since then my better hitter gets to hit instead of being walked.[/quote:0b6bd5d17a]
This is absolutely true, and it's kind of an "intangible" that is hard to test. I guess you could do another comparative study where you compared a great clutch guy hitting behind VMart to an otherwise similar lousy clutch guy, and see how Martinez's IBBs and (more importantly) team runs scored compared. Without doing that, I would definitely agree that this is one of the more important ways that clutch can affect your team. It does, however, mostly matter only when you have a team that has a couple of hitters and not much else. If you had seven good hitters, worrying about a #8 hitter who can "protect" the #7 hitter seems a little silly.

[childsmwc]

The formula to determine the impact of clutch in strat is very straight forward:

X=Clutch Adjustment
Y=Clutch PA's

Formula is: X/216*Y= clutch hits +/- over the course of a season.

The only thing in this formula that isn't a known is Y, which is what is the expected number of clutch occurrences over a typical season and then secondly can that be refined to determine the average clutch appearances for a specific batting order slot.

Now lets look at the extreme, lets say player A has 700 PA's with 175 hits and 75 walks (hits and walks really aren't relevant for the discussion however) based on the expected result of their card matched with the average pitcher used in a given league. Using Catalanatto with a +11 clutch the results would be:

11/216*700=36 so the best Catalanatto would be expected to do would be 36 additional hits. However, we know that not every PA is a clutch situation.

To determine this factor accurately you should review every PA in every game and count how many times a runner is in scoring position with 2 outs divided by total PA's on the season. I short cut this using the following methodology:

Probability that there are 2 outs: 1/3 (assuming all states of 0,1, & 2 outs are equal)

Probability that there is a man on 2b, 3b, or 2B&3B 20% (wild ass guess here)

So .33 * .2 equals roughly 6% of PA's will be clutch, but for arguments sake lets assume its 10% of the time. So going back to our original computation:

11/216*(700*10%)=3.6 hits

So that 11 clutch doesn't amount to much at the end of the day.

As you have already noted you know what Catalanotto's improved batting average in the clutch is 11/216 the 50 points you discussed. You just need to determine how many times that comes into play over a given season and to do that you would be better off counting all PA's or if you want to limit it to the #5 slot in the lineup count all teams #5 slot in the order for clutch events divided by total events.

I personally would stick with the short cut methodology laid out above,

Bbrool

[childsmwc]

The flip side to the Petco vs. Coors debate is that a park like Coors will generate more baserunners and therefore more clutch AB's over the course of a season. So while clearly a run in Petco is more important than a run in Coors, the same clutch number will actually generate more runs in Coors versus Petco so to say positive clutch is more important in a pitchers park is a misstatement when based purely on raw clutch numbers.

Bbrool

[MARCPELLETIER]

bbrool,

what you seem to forget is the difference of club make-up between Petco and Coors. In Coors, I will build my team on slugging and on-base, whereas in Petco, I will build my team on guys that walk, single, and steal (I consider bunting and hit-and-run as well for cheap hitters, so the probability that I "push" intentionally someone on second-base exists in Petco, but not in Coors). This difference, added to the fact already mentioned that 25 runs in Petco contributes to a greater W-L difference than 25 runs in Coors, together make clutch more important in Petco than in Coors.

(to use your formula, I will intentionally intend in Petco to have more Y, not in Coors, both by coaching decision ---increase bunt and H-R with cheap hitters---and by players selection---players that singles and walk).

Dean,

to express bbrool's point in English rather than in Maths, even though your formula ended up with a greater correlation for 9th spot than for 4th spot, in absolute numbers---in actual numbers of plater appearance, the 4th spot probably see more clutch situations nonetheless. It is therefore my "gut feeling" than 5th, than 4th, are the two important spots to consider for clutch. This last argument is also an argument to place your best slugging players in both 4th and 5th spots.

That said, in a Petco team that I just built, I don't hesitate to use Duffy at least at home (whose card not only is clutch, but is full of single**), even though he has no power. Two other reasons, other clutch, push me to do so: I have only one slugger in this team---Cabrera who is batting 4th. By having Duffy, I force my opponents to face Cabrera. Also, my only other "slugger" is Johnson---not your typical strong slugger.

Three games in the books so far, and Duffy has already two rbi singles in clutch situations, including in a one-run win. 8)

Don't forget that in National leagues, the numbers are skewed differently, with #1 having more clutch situations (anytime the pitcher bunts with one out).

[childsmwc]

You know as I thought more about why is clutch more important in a pitchers park versus a hitters park, something bothered me. And then I realized we were assuming that since a run in a pitchers park amounts to a bigger increase in output relative to the baseline versus that same run in a hitters park, it is therefore more important. However, we have failed to consider the relative value of runs.

If in a pitchers park my average score is 4.5 to 4, so on average I am outscoring my opponents by .5 run then that same lineup in a hitters park might have an average score of 5.5 to 5 and on average still outscoring opponents by .5 runs a game.

In that example why would a clutch run in the pitchers park be worth any more than in a hitters park. My average run differential is identical which implies that a run has the same benefit to the outcome of the typical game.

It would seem that the only way a run is more valuable in a pitchers park is if you run differential is say .5 runs a game there yet say 1 run a game in a hitters park. If the two parks produce different expected run differentials then 1 clutch run in the ballpark with a smaller run differential would have more impact. However, since run differential has a direct correlation to winning, and if there was a fundamental difference in expected run differentials, owners would only use big ballparks. However, that is not the case.

It would seem to be implied through owner behavior that a pitchers park produces a hypothetical 4.5 to 4.0 game while a hitters park produces a hypothetical 5.5 to 5.0 game. Same result, just different means of getting there and thus clutch impacts both environments the same.

Bbrool