The importance of ballpark: more to it than BPs

The importance of ballpark: more to it than BPs

Postby MARCPELLETIER » Fri Oct 07, 2005 3:17 pm

I always had great success in Coors, but had inconsistent success in stadiums like Petco, so I started a few months ago to think why this was the case. After all, we all know about the importance of drafting a good line-up that fits your stadium. If you go with Petco, you select offensive players with few ball parks homeruns (say Snow), and you don't hesitate to select pitchers with substantial ball park homeruns, as long as they have low on-base and relative low total base other than BPs. Let's call this the basic stadium principle. I always respected this prinicple, so why did I struggle in Petco?

It is a few months ago when I realized that my failures could be explained by another principle that I was not considering. According to the recent sabermetric studies (well not so recent), the value of events---the value of a walk, for example---really depends on the context in which it happens. The context is determined by the situation (is it bases loaded or bases empty) and also by the stadium.

One very intuitive example easy to understand is the context of empty bases. In this context, a walk is as good as a single: There is no one to advance, and the guy will end up at first anyway.

(By itself, this example hinders an often overlooked principle: it is not necessary the player with the most on-base that should lie on top of the line-up, it also depends on the composition of walks and singles on the cards. It is better, from an effecient point of view, to have a player with more on-base coming exclusively from hits in the second or third spot of the line-up and have a lead-off with a bit less of on-base, but whose on-base comes mostly from walks. Of course, this is true only for dh-leagues. With pitchers bunting, it might be a good idea to have a hitter leading off).

In which way do stadium ratings influence the context? The response is in modifying the easiness to have an out. It would be too long to explain how it works out (I'll do so in another post if someone is interested), but bottom-line is this: the easier the out, the greater the difference between singles and walks. The tougher the out, the smaller the difference between singles and walks. Consequently, in Coors, where outs are difficult to get, it is avantageous to draft players who walk a lot (call it the walkers) over players who single a lot (call it the hitters), whereas in Petco, hitters should be preferred to walkers. Similarly, pitchers who give lots of walks should be preferred to pitchers who gives lots of singles, other things being equal, of course.

(One note: of course, if you believe that TSN undervalues walks in general, it might be advantageous to select walkers over hitters regardless of the parks, but of course, this does not say anything about the truth of the context principles underlined here. Personally, I still think that walks are slightly undervalued in TSN, but I think they are valued fairly enough to have hitters worth more than walkers in stadiums like Petco).

Another context principle related to stadiums is that homeruns have more importance with regards to the other slugging hits (doubles-triples) in Petco rather than in Coors. Of course, this goes against the basic stadium principle that you should avoid homerun hitters in Petco, and the basic stadium principle has more impact on the efficiency of a team than the context principle just seen. Taken together, they make the third principle: when playing in Petco, you should avoid offensive sluggers with lots of BPs, but for the one or two sluggers that will make your team, go for those with the most direct homeruns.

These ideas have been thrown in these boards in the past, but I thought it would be neat to summarize them together here. Furthermore, I drafted my last team entirely following these principles.

I drafted/picked up two walkers: 2b Jimenez, and 1b Snow. I picked up Jimenez after I traded away Hudson for better pitching, with the idea of making Jimenez my lead-off vs rhp. I drafted Snow according to the basic stadium principle---he was the best premier offensive non-sluggish first basemen available. These are the two "walkers". All my other players are basically hitters.

My one slugger is Beltre. Tons of direct homeruns vs rhp.

All other offensive players were typical hitters. They all have the same profile, basically: .320, batting average; .360 on-base, with few BP homers. They are c-Burke, ss-Izturis, lf-Rivera, cf-Kotsay, rf- Suzuki, and dh- Newhan. Because this is a 100K team, I had extra dollars that I spent on Sexson (walkers with lots of BP, useful when I visit Coors fields) and Marcus Giles (to be second base vs lhp).

As for pitching, I drafted/picked-up Sheets (3BP vs lhp) as my ace. Leiter and Harden were chosen because they allow a good number of walks, but few other things. The rest of my pitching squad have a lot of BPs.

My record is 32-22, which is good, but not extraordinary. My pythogerian record though is 37-17, and I am first in both runs scored and runs allowed. The reason why I underperformed is because I didn't respect a fourth principle: when playing in Petco, make sure you have a dominant closer!! My comittee of closers (Aquino, Smoltz) had six blown saves before I could even say "scissors". Hopefully, the coming of Benitez will stabilize my team in late-inning games.

Of course, because my team is playing in Petco, I don't expect it to finish first in offense. Surely this is an outlier. Nevertheless, it is remarkable to notice that my team is first despite being 11th in hr, 11th in doubles, 10th in walks, 11th in slugging---not an outlier anywhere there. However, my team leads the league by far in hits, particularly in singles, so much that my first is first---but by a small margin--in on-base.

I thought that one reason why my team played so well offensively is because it was lucky in the clutch. But after 54 games, my team is hitting .285 and slugging .415 in the clutch, a few points below the .287/.419 mark overall.

One outlier is Beltre's rbi production: 54 rbis in 54 games. Yet, his stats don't look all that good: .289/.335/.589, not even in the top categories for runs created. He wasn't particularly clutch either. He just seems to hit at the right place, behind walker Jimenez and hitter Suzuki.

There is one other thing that I have changed from my usual Coors way of managing my team, but I won't tell it now. I let you guess. :wink:

As we speak, I don't know if I will be able to maintain the success of my team, but I am now convinced more than ever that there is more than BPs to consider when it's time to fit your roster with your stadium.
Posts: 55
Joined: Tue Jul 03, 2012 2:34 pm

Postby mfsleeze » Fri Oct 07, 2005 4:07 pm

I oversimplify because I don't have the time to do the depth of analysis you are doing, but I agree with your approach. I basically value TB over OB as my measure of who to draft in pitchers' parks, where in hitters' parks I'm much more interested in OB.

Also, Justin Morneau is a good example of your 3rd principle. He has max BPHR chances, but also the same direct HR chances vs RHP that David Ortiz has. But I think because of the BPHRs most people would shy away from him in a pitcher's park.
Posts: 55
Joined: Tue Jul 03, 2012 2:34 pm

Postby J-Pav » Fri Oct 07, 2005 10:05 pm


Great post! Also, it's good to see Leiter finally getting love north o' the border :wink:

Good luck the rest of the way...

Posts: 55
Joined: Tue Jul 03, 2012 2:34 pm

Re: The importance of ballpark: more to it than BPs

Postby cummings2 » Sun Oct 09, 2005 12:27 pm

[quote:6f69045905="marcus wilby"]In which way do stadium ratings influence the context? The response is in modifying the easiness to have an out. It would be too long to explain how it works out (I'll do so in another post if someone is interested)[/quote:6f69045905]

I'd be quite intersted in that post :D I understand the "bottom line" (at least I think I do) but I'd be interested in HOW it works out...

So far I have focused my learning in strat to understanding pitching and defense, my offense however has gottten the nasty side of my stick, I hope this post helps my future teams.

I have a Petco team that has been a disaster so far, when building the offense I did so with lefty hitters who would most likely use the only offensive "advantage" that Petco gives:LHB Singles. I understand it isn't really an "advantage" but the way I figured it my LHB's AVG wouldn't dip as much (and my OBP either) and the offense would be able to score just enough runs to generate some wins. I ended up with the likes of Erstad and Dejesus, Nixon and some power guys that I thought would provide power mostly out of Petco...failed strategy: my team AVG is .230 and my OBP in .295.

I should add when it comes to the offense that I thought that in order to maximize the effect of Petco on my pitching I should have the best possible defense so I sacrifized 3 positions 2B -Pokey Reese. SS -Adam Everett and CF -Steve Finley for the best possible (and available) defense. I knew they would struggle with the bat...I just didn't think they wouldn't be able to hit .200 after 114 games. My logic behind that move was that the stadium provides an edge to the pitching staffs (home and away) so what the home team can do to have a distinct difference over the visiting team that fits the stadium's "personality" is on Defense and LHB in the lineups, I thought it more likely for a hit to challenge my defensive ratings that my BP ratings...what I hadn't taken into consideration is the EASINESS TO HAVE AN OUT! (If my 2B gives a one base error with the bases empty and 2 outs it is less likely to turn into a 2 run jimmy jack than in Coors :) O.K. I think I get this...

Thanks Luckyman!
Posts: 55
Joined: Tue Jul 03, 2012 2:34 pm

Postby MARCPELLETIER » Sun Oct 09, 2005 7:50 pm




the edge for singles to lh is so little that, personnally, I barely take it into account. Much more important is to consider the lack of BPHR and, of course, to have a line-up which maximizes the weakness of your opponents' pitching squad (have reverse lh if your opponents have reverse lhp, etc).

Which leads me to a caution. To have success, there are more important steps to consider than the one described here. What I stress here is to make an extra step---in composing an offensive line-up---into a process that presupposes many more prerequisite steps.

To explain how a specific environment influence the value assigned to hits and walks, I will take two examples.

First, suppose you are facing Pedro Martinez in his better days and that you know in advance that you will have ONE hit, no walk, no SB, no free bases from errors. In this simulation, you really hope that the one hit will be a hr. Indeed, you can pretty much be certain that, if this hit is a hr, you will score one run, and only one run, and if this hit is a single, chances are you will be shut out. In this environment, a hr is worth 1 run and a single is worth virtually 0 run. In this very specific environment, a single is almost no different than an out.

LEt's consider now the opposite situation. Suppose you know in advance that your team will have 15 consecutive players reach safely on base in the first inning of your next Coors game. In this context, you don't really care if your players get walks, hit singles, or hit homeruns. If your team hits 15 straight hr, then you will score 15 runs; if, on the other hand, your team has 15 straight walks, then you will score a minimum of 12 runs, and perhaps more, depending on what type of outs follow. It is like that 13 or 14 of these players will turn in to score a run. In this very particular context where you don't encounter any out, all events, walks, singles, doubles, are all worth a value close to 1 run, just like hrs. In this context, singles is very different from an out, but not so different than a hr.

These two situations represent the extreme situations of the following graph, which represent the specific value of hits and walks, depending on the context:

At the left, you have the example with Pedro--an environment with an obp close to zero. At the right, you have the example of an environment where there is virtually no out. In real life, the OBA oscillates between .300 (for Petco environments) to .400 (for Coors environements). This clearly shows that, in Petco environments, doubles and triples relative to hr lose a lot of value. That being said, in Petco you lose your BPHR as well. The net result of this is that, while you need to avoid players with lots of BPHR, when you have to take one slugger, you have to consider players with a lot of direct HRs. On the opposite, in environments where you also lose your BPHR but where on-base is high--SBC Park, Fenway, particularly when you have very good lead-offs...--- a good strategy is to turn to players with lots of doubles and triples.

In this graph, it's not clear that there is a difference between walks and singles in .300-OBA environments as opposed to .400-OBA environments. This is because, with bases empty, a walk is worth as much as a single, and empty bases are likely to happen more often in .300-OBA environments than in .400-OBA environments. When this is factored in, then, for the fewer occasions when there are men on base, the single has a relative weight to walk that is more important in .300-OBA environment than in .400-OBA environment. And this is particularly true if you are more aggressive running around the bases.
Posts: 55
Joined: Tue Jul 03, 2012 2:34 pm

Postby cummings2 » Sun Oct 09, 2005 9:05 pm

Okey I need time to digest the info and run some numbers just to make sure I get it...I think that to a limited extent I have used some of this principles (on sheer instict) when building Safeco teams and choosing my RH power sources but that's a diferent post.

Now a question:

I am quite new to Strat and so far I am focusing mostly on understanding pitching and defense. I have been toying with a formula that would help me estimate OAVG (L & R) vs my pitchers in order not only to draft who I consider the best fit but also when a player is actually under/over performing, that is to give me and idea of when I should be patient and wait for the pitcher to turn it around or when I made a mistake in selecting the player and when I should cut my losses. So far it is proving to be fairly accurate but there are a number of questions that I have so as to make it more dependable and accurate.

The question is: Is there a formula out there that already does this so I can compare it to what I am doing and learn from my "mistaken logic", if there is where I can I find it and if it isn't out there is someone interested in going through it with me.

But anyway, back to the ACTUAL topic of this board, thanks a lot Luckyman, will go through the numbers and try to learn a bit more.

If you don't mind my asking. I do have a question about the Petco team you used as an example above:

What exactly do you mean by "my Pythagorean record"

Thanks again for the graph, the help, the extreme examples made it WAY easier to understand... and continued good luck!
Last edited by cummings2 on Sun Oct 09, 2005 10:20 pm, edited 2 times in total.
Posts: 55
Joined: Tue Jul 03, 2012 2:34 pm

Particularily impressed by...

Postby cummings2 » Sun Oct 09, 2005 9:15 pm

forgot to mention, I'm particularily impressed by the number of runs your team has generated AT HOME, at first I thought you were getting a lot of those runs that have you in 1st on RS AWAY from home but you have consistently scored high numbers in petco (149R out of 311 runs scored so far)...if you don't mind I'll take a closer look at the inning break downs of each game to learn a bit more about this whole issue.

Again, All best of luck
Posts: 55
Joined: Tue Jul 03, 2012 2:34 pm

Postby MARCPELLETIER » Sun Oct 09, 2005 10:41 pm

Formula for expected OAVG: a bit complicated given that you have to consider hits from stadium, defense, and offensive and defensive cards. That being said, if the purpose is to determine whether you should keep a pitcher or take the 20% cut, the answer (for non 1980's leagues) is easy: never take the cut, unless the pitcher is cheap (2M-max, I would say).

Pythagorian record is the record you "should" have given the number of runs you have scored and allowed. It's basically:

Run scored^1.83 / (Runs scored^1.83 + runs allowed^1.83).
Posts: 55
Joined: Tue Jul 03, 2012 2:34 pm


Postby Leo / loob » Sun Oct 09, 2005 10:49 pm

Please take a look at this Petco team. Not sure if it proves, or, disproves your theories, but, in the early stages, it sure is kicking arse....

After 18 games (9 home and away)
League Hitting Averages
BA: .258
SLUG %: .430
OBP %: .334

Team Hitting Averages
BA: .284
SLUG %: .433
OBP %: .339

League Pitching Averages
ERA: 4.48
WHIP: 1.41

Team Hitting Averages
ERA 3.58
WHIP 1.26

Runs scored at home 42
Runs scored away 45
Leo / loob
Posts: 55
Joined: Tue Jul 03, 2012 2:34 pm

Postby MARCPELLETIER » Sun Oct 09, 2005 11:16 pm

18 games (or 60 games) is waaaaaaaaaaaayyyyyyyyyyyyyy too few to conclude anything.

Fact of the matter is, sometimes 162 games ain't enough.

Hoping your success will continue.
Posts: 55
Joined: Tue Jul 03, 2012 2:34 pm


Return to Strategy

Who is online

Users browsing this forum: No registered users and 3 guests