So far, we’ve looked at seeing the overall impact be the same, but for different reasons, like the bases loaded walk versus the solo HR. And you are either given a choice of “same” or “this one is better”. It’s been designed to see if you prefer a DIRECTION. In the above case, the direction is based on whether the event matters.
Now, we’ll have two situations that have an overall different impact. We have our trusty solo HR. We all love the HR. It gives us a guaranteed run. It tells us about the hitter, and if given better circumstances, we can dream of even more runs.
If I ask about a double with a runner on second, I’d get into the same question as with the bases loaded walk: the impact is exactly one run, and we’re left with the same state that we entered. And 80% of Fangraphs readers will prefer the solo HR to the double with the runner on 2nd.
But how about a bases-clearing double? Any way you want to measure it, the impact is going to be alot(*) more than the 1 run from the solo HR. Run expectancy tables tell us it’s 1.7 to 2.2 runs depending on the number of outs.
(*) “Get over it.” — Scalia
So, how much of a hurdle are Fangraphs readers willing to climb to keep their allegiance to the event in a context-neutral setting, and ignore the context of a somewhat inferior event, but in a highly more leverageable setting?
You Aren't a FanGraphs Member
It looks like you aren't yet a FanGraphs Member (or aren't logged in). We aren't mad, just disappointed.
We get it. You want to read this article. But before we let you get back to it, we'd like to point out a few of the good reasons why you should become a Member.
1. Ad Free viewing! We won't bug you with this ad, or any other.
2. Unlimited articles! Non-Members only get to read 10 free articles a month. Members never get cut off.
3. Dark mode and Classic mode!
4. Custom player page dashboards! Choose the player cards you want, in the order you want them.
5. One-click data exports! Export our projections and leaderboards for your personal projects.
6. Remove the photos on the home page! (Honestly, this doesn't sound so great to us, but some people wanted it, and we like to give our Members what they want.)
7. Even more Steamer projections! We have handedness, percentile, and context neutral projections available for Members only.
8. Get FanGraphs Walk-Off, a customized year end review! Find out exactly how you used FanGraphs this year, and how that compares to other Members. Don't be a victim of FOMO.
9. A weekly mailbag column, exclusively for Members.
10. Help support FanGraphs and our entire staff! Our Members provide us with critical resources to improve the site and deliver new features!
We hope you'll consider a Membership today, for yourself or as a gift! And we realize this has been an awfully long sales pitch, so we've also removed all the other ads in this article. We didn't want to overdo it.
Gosh, I am almost tempted to abstain because this comparison seems unfair (and of course it’s intended that way). The discrete partial run values of each event that created the runners on base has been credited right? Do we give double credit for timing those events well or give timing credit to a hitter put in a clutch situation? I think I am OK with there being a sequencing bucket that is not attributable to any individual players because it takes a team to sequence.
MattabattacolaMember since 2020
10 years ago
I think first WAR should be calculated completely context neutral. A player with 40 doubles with lots of guys on base (so maybe 100+ RBI’s) should not be as good as the guy who hits 40 homers on a bad team. Once you take all the context neutral data, then its possible to compare situations. For example, stats like batting average with runners is scoring position, or maybe something like…
Runs produced/run expectancy
Basically if you were to keep this stat for every situation you’d see that players who outperform their expectancy’s would have a higher number. It would allow us to get a snapshot of how they performed.
SabrRattling
10 years ago
Is it weird for me to believe batters should be context neutral, as they have no control over what their fellow batters do, but to think some context needs to be used with pitchers because they control every at bat?
I think the only time context might be considered for the batter is high leverage situations – and only minorly.
For pitchers, I think some sort adjustment needs to try and account for defensive errors or shortcomings (i.e. Manny).
I could be thinking about this all wrong, but obviously some people here agree based on the way voting has turned out so far. I could also be colored too strongly by my playing experience where I definitely pitched differently based on the score/situation. Granted I also batted differently in certain situations.
I’d also be curious if opponent quality could be factored in somehow. Like a bases clearing double off Wade Davis (with Dyson in right) vs a solo hr off Ian Kennedy.
/apologies if covered in previous comments
Phillies113
10 years ago
The home run wins, even if the double cleared the bases. The batter had nothing at all to do with the previous batters reaching base; why should he receive credit for hitting a double just because they were skillful (or lucky) enough to reach base? I’m sticking to my context-neutral guns here.
The only way, really, I can see myself considering the context is in the way the pitcher approaches the batter. Depending on the batter, the pitcher, and the base-out state, the pitcher may choose a more aggressive approach, attacking the strike zone and daring the batter to swing. Or, the pitcher can be more conservative, pitching around the edge of the zone to fool the hitter into swinging at junk.
Still, the batter has to have the wherewithal and the patience to know when to swing and when not to swing. In those instances, I’m not sure context really matters too much; batter sees a hittable ball, batter hits the ball. Context is important for narrative and for generating wins, but we’re talking WAR here. We want to evaluate the batter’s ability to bat. I just don’t think context is as important when making such an evaluation.
Adam Maahs
10 years ago
A thought that might satisfy the divide between game impact and context neutrality:
The distribution of possible outcomes and their impact is laid out in the calculation of the leverage index. A player can’t control the leverage of the situation he is thrust into (and therefore the spread in the win expectancy change of the outcomes) but he can control the outcome he generates.
What if we normalized the list of outcomes so the worst possible outcome (strikeout, triple play, etc.) sits at zero and the best possible outcome (usually a HR but maybe tied with a hit or walk in extreme cases) is set to 1. All other outcomes fall somewhere in between.
A player could be judged on a per at bat basis to determine their average outcome between 0 and 1, a nice familiar scale and something that could be substituted for wOBA for an alternate WAR system. I suppose this is kind of similar to WPA/LI.
Interested to hear anyone’s thoughts on this
Bernie Sanders
10 years ago
Is there any way we can spread the wins around to everyone around the league. It doesn’t have to be shared equally–after all, they don’t call me an admitted democratic socialist because I’m an admitted authoritarian socialist–but it seems wrong that players like Skip Schumaker end up with negative wins every year. Surely you wouldn’t claim it was humane what happened to Adam Dunn in 2011?
We need to be looking after the Dee P. Gordon’s of the world, the people working 168 hours per week through a torn labrum, a herniated disc, a sore wrist, biceps tendinitis, a forearm strain, two damaged elbow ligaments, two mild hamstring strains, a sore ankle, a sore forearm, acute elbow soreness, a sore thumb, a strained hamstring, a wild calf strain, an intercostal strain, and a tired arm because their employers don’t allow its workers to take time off!
Bad Hermit
10 years ago
Is some of this context-neutral talk being muddled with intent-consistency?
Bad Hermit
10 years ago
I may be ignorant of this already being addressed, but I also find myself wondering about the following:
Assuming the bases-clearing double is favored to the solo HR, then two players have 600 PAs.
Player 1 faced 600 PAs of the polled bases-loaded question. In these situations, Player 1 hit the bases-clearing double 50% of the time, and Player 1 struck out the other 50% of the time.
Player 2 faced 300 PAs of the polled bases-loaded question. In these situations, Player 2 hit the bases-clearing double 50% of the time, and Player 2 struck out the other 50% of the time.
Player 2 also faced 300 PAs with the bases empty. In these situations, Player 2 hit a double 50% of the time, and Player 2 struck out the other 50% of the time.
Do both Player 1 and Player 2 accumulate the same amount of context-considered value from their respective 600 and 300 PAs with the bases loaded?
Gosh, I am almost tempted to abstain because this comparison seems unfair (and of course it’s intended that way). The discrete partial run values of each event that created the runners on base has been credited right? Do we give double credit for timing those events well or give timing credit to a hitter put in a clutch situation? I think I am OK with there being a sequencing bucket that is not attributable to any individual players because it takes a team to sequence.
I think first WAR should be calculated completely context neutral. A player with 40 doubles with lots of guys on base (so maybe 100+ RBI’s) should not be as good as the guy who hits 40 homers on a bad team. Once you take all the context neutral data, then its possible to compare situations. For example, stats like batting average with runners is scoring position, or maybe something like…
Runs produced/run expectancy
Basically if you were to keep this stat for every situation you’d see that players who outperform their expectancy’s would have a higher number. It would allow us to get a snapshot of how they performed.
Is it weird for me to believe batters should be context neutral, as they have no control over what their fellow batters do, but to think some context needs to be used with pitchers because they control every at bat?
I think the only time context might be considered for the batter is high leverage situations – and only minorly.
For pitchers, I think some sort adjustment needs to try and account for defensive errors or shortcomings (i.e. Manny).
I could be thinking about this all wrong, but obviously some people here agree based on the way voting has turned out so far. I could also be colored too strongly by my playing experience where I definitely pitched differently based on the score/situation. Granted I also batted differently in certain situations.
I’d also be curious if opponent quality could be factored in somehow. Like a bases clearing double off Wade Davis (with Dyson in right) vs a solo hr off Ian Kennedy.
/apologies if covered in previous comments
The home run wins, even if the double cleared the bases. The batter had nothing at all to do with the previous batters reaching base; why should he receive credit for hitting a double just because they were skillful (or lucky) enough to reach base? I’m sticking to my context-neutral guns here.
The only way, really, I can see myself considering the context is in the way the pitcher approaches the batter. Depending on the batter, the pitcher, and the base-out state, the pitcher may choose a more aggressive approach, attacking the strike zone and daring the batter to swing. Or, the pitcher can be more conservative, pitching around the edge of the zone to fool the hitter into swinging at junk.
Still, the batter has to have the wherewithal and the patience to know when to swing and when not to swing. In those instances, I’m not sure context really matters too much; batter sees a hittable ball, batter hits the ball. Context is important for narrative and for generating wins, but we’re talking WAR here. We want to evaluate the batter’s ability to bat. I just don’t think context is as important when making such an evaluation.
A thought that might satisfy the divide between game impact and context neutrality:
The distribution of possible outcomes and their impact is laid out in the calculation of the leverage index. A player can’t control the leverage of the situation he is thrust into (and therefore the spread in the win expectancy change of the outcomes) but he can control the outcome he generates.
What if we normalized the list of outcomes so the worst possible outcome (strikeout, triple play, etc.) sits at zero and the best possible outcome (usually a HR but maybe tied with a hit or walk in extreme cases) is set to 1. All other outcomes fall somewhere in between.
A player could be judged on a per at bat basis to determine their average outcome between 0 and 1, a nice familiar scale and something that could be substituted for wOBA for an alternate WAR system. I suppose this is kind of similar to WPA/LI.
Interested to hear anyone’s thoughts on this
Is there any way we can spread the wins around to everyone around the league. It doesn’t have to be shared equally–after all, they don’t call me an admitted democratic socialist because I’m an admitted authoritarian socialist–but it seems wrong that players like Skip Schumaker end up with negative wins every year. Surely you wouldn’t claim it was humane what happened to Adam Dunn in 2011?
We need to be looking after the Dee P. Gordon’s of the world, the people working 168 hours per week through a torn labrum, a herniated disc, a sore wrist, biceps tendinitis, a forearm strain, two damaged elbow ligaments, two mild hamstring strains, a sore ankle, a sore forearm, acute elbow soreness, a sore thumb, a strained hamstring, a wild calf strain, an intercostal strain, and a tired arm because their employers don’t allow its workers to take time off!
Is some of this context-neutral talk being muddled with intent-consistency?
I may be ignorant of this already being addressed, but I also find myself wondering about the following:
Assuming the bases-clearing double is favored to the solo HR, then two players have 600 PAs.
Player 1 faced 600 PAs of the polled bases-loaded question. In these situations, Player 1 hit the bases-clearing double 50% of the time, and Player 1 struck out the other 50% of the time.
Player 2 faced 300 PAs of the polled bases-loaded question. In these situations, Player 2 hit the bases-clearing double 50% of the time, and Player 2 struck out the other 50% of the time.
Player 2 also faced 300 PAs with the bases empty. In these situations, Player 2 hit a double 50% of the time, and Player 2 struck out the other 50% of the time.
Do both Player 1 and Player 2 accumulate the same amount of context-considered value from their respective 600 and 300 PAs with the bases loaded?