Agree or disagree and why?

Preventing walks is more beneficial in offensive parks, while getting walked is more beneficial in pitchers parks.

Hmmm...first question is "more beneficial than what?" I think you mean "preventing walks is more beneficial in offensive parks than in pitchers parks, while getting walked is more beneficial in pitchers parks than in offensive parks."

If this is what you meant, then I think the statement is contradictory. If (as I believe) the marginal run value of a walk is higher in a pitchers park, then that run value is the same whether viewed from the perspective of the offense or the defense. So if getting walks is more beneficial in pitchers parks, that would mean preventing them is more beneficial there as well.

If you see it differently, I'd love to understand the math, or at least the theoretical reasoning, behind your view. Why would the run value of a walk, compared to other offensive events, be different for the offense than for the defense?

I haven't done the math so you may have better numbers than I do here. Here's my thought.

In a pitcher's park, baserunners are more rare in general, so the run-value is higher and therefore more enticing to the hitting team that it otherwise would be, as you stated above.

But, in a hitter's park, a walked batter is inherently more likely to score since the next few batters up are more likely to get hits than they would be in a pitchers park. Therefore, it is more important (from the pitching team's perspective) to not allow a walk in the first place than it would be in a pitcher's park.

I was thinking about this as I was building a team for Coors Field, and I remembered that most people recommend focusing on OAV metrics for their pitchers in Coors, but I almost always focus on BB/9 metrics. I had never really thought much about it before and just kind of did it because, from a cost per incremental baserunner allowed perspective, it made sense to focus on something I could have more impact on (i.e. spending the same amount on better OAV would result in less of a benefit in Coors, since the "hit multiplier" was higher. Tough to explain without showing the math, and I may be off base, but that's my thought at least. It's the same reason why paying for defense and range is so essential in Coors, while it isn't really as important in Safeco.

Walk rates and strikeout rates are going to be consistent regardless of what park you play in, so the relative variable that will actually change is Avg, or more specifically BABIP. For that reason I skew towards low OAV pitchers/high Avg-high contact hitters in parks like Hilltop, and low BB pitchers/high BB hitters in parks like Petco.

I've never done it but I'd love to hear if anybody has tested a high k/9 pitching staff in a hitter's park - like Waddell, Dean, Feller, Ryan etc. I've actually gone the other way and used low k/9 and bb/9 guys like Babe Adams and Cy Young in Petco and it works out very well.

A second question - what threshold of bb/9# do you guys max out at? I've struggled with this - I try not to have starting pitchers north of 3.0 and I will dabble with relievers as high as 3.5 bb/9, but anything north of that usually has bad results, especially when you run into high-OBP teams.

It generally depends on the restrictiveness of the theme, the other qualities of the pitcher and the salary cap but, in general, I max out somewhere between 2 and 2.5

Also, the idea that walk rates will remain the same despite avg changing is precisely why the importance of a walk would change.

You have to save $ somewhere. If I'm in Coors I want OAV at any cost, walks be damned. Maybe that thinking is wrong.

Don't you factor the league competition somewhat, too? If you know you'll be facing a bunch of Madduxes who issue few walks regardless, maybe it's not worth spending to try to get more of them. Get the higher average hitters in, since they'll maximize what the pitcher allows?

I think redcped has a good point, but let's put that aside from now...all else being equal, without even getting into the math, I still think the original proposition is inherently a contradiction. The run value of any offensive event is the same whether viewed from the standpoint of the offense or the defense. So if it's more valuable to get a walk in a certain environment, then by definition I think it has to be more valuable to prevent a walk in that environment. Unless I'm really missing something.

Yeah, competitiveness is definitely a factor -- I tried to capture that by mentioning "restrictiveness of the league" but I suppose they are actually two different things.

How are you defining run value, contrarian? Maybe we are using different definitions.

When we think about the likelihood of someone on first base ending up scoring, it seems intuitive that it would be higher in hitter's parks. So, keeping someone from getting there is more valuable to the team pitching. But, because it naturally will happen more often (due to there being more hits), the incremental value -- or, importance -- of making it happen (via a walk, in this example) would actually be lower for the hitting team in a hitter's park than a pitcher's park.

If we want to get into the math, though, this chart might be useful:

http://www.tangotiger.net/customlwts.html

It basically shows how the run value of different events in different contexts (high-scoring, low-scoring, etc.)

Basically, the run value of EVERY positive event increases in higher run scoring environments.

What I find interesting - and I confess I did not expect this - is that it appears the RELATIVE value of a walk, compared to a single, INCREASES in higher run contexts.

In a one run per game environment (not realistic, I know, but just to set a baseline), a walk is worth .144 runs, and a single is worth .238. So a walk is worth .144/.238 of a single, or 60.5%. In 4 run environment, it's 65.9%, in a 6 run environment it's 67.8, and in an 8 run environment it's 69.4%.

I'm still not 100% sure how to translate this into WIS strategy, but it seems to suggest that all else being equal preventing a single is more important than preventing a walk in low-run contexts. And while that continues to be true in higher run environments, the gap closes. Does this mean that given two pitchers with comparable WHIP and comparable HR/9, we should give more weight to low OAV in pitchers parks and more weight to low BB in hitters parks?

Curious what others think?

I would guess, and this is just a guess, that in an environment with fewer runs, there are also fewer hits, and the walks are worth less because the runners never score. And in the enviroment with more runs, every PA is more likely to produce something good.

So while the value of a single run is less in a high-scoring environment, the value of a single baserunner/PA is higher because something is more likely to come of it.

Think back to Moneyball - the goal is avoiding outs. If your average hitter hits 290/370/480, you really want that next guy at the plate. If your average hitter hits 230/300/380, then go ahead and walk people, the next guy is getting out anyway.

A single is worth more because it can advance another runner more than one base. With nobody on base, a walk and a run are worth the same in expected value.

For purposes of WIS, a walk is worth more in a pitcher’s park because of the $ value that is attributed to the hitter and your overall resource allocation. You are paying for some combination of runs created when you buy a player - that is a function of his OBP and slugging. So you are effectively paying for some portion of that OBP as determined by two variables - walk and hit rate. In the pitcher’s park, that hit rate is devalued while walk rate is constant, so the manager that spends his $ on players with a higher walk rate to begin with will be extracting more runs created for his $ spent. It is the opposite for a hitter’s park (and likewise you can extrapolate this $ allocation strategies to pitchers as well).

Here’s a visual way of looking at it...imagine two players in a park that decreases hits by 20% (roughly Petco in my experience)
Player 1 .200/.400/.500 —> in petco .160/.360/.460
Player 2 .300/.400/.500 —> in petco .240/.340/.440
- these two players initially have the same runs created (at least by the simple mathematical equation) and will theoretically be priced the same. In Petco, however, Player 2’s runs created falls well below, as the decreased hit rate slams his OBP and SLG; Player 2 would also be considerably more valuable in the hitter’s park that increases hit rate by 20%
- this isn’t perfect because for purposes of the post I’m assuming they’re losing only singles, but hopefully you get my point

All else equal (aka a neutral park), I would skew towards having higher average hitters, because 1) hits are worth more, as described above and 2) you can’t control the pitchers you go against and if you go with high bb hitters and run into a no-walk Babe Adams, you’re SOL.

Posted by jfranco77 on 1/10/2019 5:00:00 PM (view original):
I would guess, and this is just a guess, that in an environment with fewer runs, there are also fewer hits, and the walks are worth less because the runners never score. And in the enviroment with more runs, every PA is more likely to produce something good.

So while the value of a single run is less in a high-scoring environment, the value of a single baserunner/PA is higher because something is more likely to come of it.

Think back to Moneyball - the goal is avoiding outs. If your average hitter hits 290/370/480, you really want that next guy at the plate. If your average hitter hits 230/300/380, then go ahead and walk people, the next guy is getting out anyway.

I think the question is whether hitter 1 at 300/410/500 is worth more than hitter 2 at 330/380/480, assuming their defense and park effects are equal. In real life, hitter 1 is more valuable. In the sim, I think hitter 2 is more valuable because of the ability to play in parks with strong + effects for hits.

In the early days of the sim, it was possible to win with offenses based on walks and power. It's not possible anymore in most leagues. Hitter 1 would be more valuable offensively IRL than hitter 2 but I don't believe that it works that way in the sim.