For those of you who bristle at sabermetric awards in historical sims, this is for you

[uruguru]

Just for the record, I am a big advocate for sabermetrics! But in my historical sims, seeing players receive awards clearly based on the sabermetric analysis within the game sort of breaks immersion for me. I'm not particularly interested in the most statistically correct player winning awards... this is just a game, so no one is being robbed imo.

I'm more interested in awards being assigned in ways that seem historically accurate given the types of analysis available at the time being simmed. If this seems crazy to you, ignore this post. But if your curiosity is piqued, read on.

I let the AI give all of the awards except MVP and CY. The MVP in my sim is called "The Bambino" and is an offense-only award. Pitchers cannot win it, they have their own award. And relievers can't win the Cy... they also have their own award.


1) The MVP. The goal is not to suggest these are statistically optimal strategies, but they give pretty good historical picks.

I add up RBI+R+AVG+SB... the highest total wins. I start by listing these four values for the RBI leader and then look for players who are comparable. RBI has a large range and variability, so the RBI leader is generally going to be in the running. I then calculate the deltas for other close players. I don't ever add up the 4 statistics. Those numbers are big. I just add up the deltas to the RBI leader.

for example, here is my "worksheet" for this past season:

mantle 289+123+123+17
kaline +26 -14 -1 +11 = +22

mays 326+132+117+ 10
robinson +5 -18 +10 +13 = +10

The AI wanted to give the MVP to Mantle & Mays -- two centerfielders with probably some defensive WAR factored in. I gave them to Kaline an Robinson, which the AI had listed each as 2nd. Everyone else was clearly out of the running because the RBI leaders are typically going to be very high on this list.

Why add SB? Because if you have a Maury Wills type season, you are going to get the attention of voters

If there is a tie-breaker, most home runs wins.

2) The CY

This award looks at only 3 stats... wins, losses and ERA

I start with the ERA leaders and then look for other good ERA starters and then adjust their ERA up or down by .1 runs for every win there are under or over the leader, and then up or down by .05 runs for every loss they are over or under the leader. Lowest adjusted ERA is the CY. If the ERA leader is leading in wins, he's almost certainly a lock (unless he has a ton of losses).

The "worksheet" for my last season looked like this (ERA leaders had high wins, too)

downing 19 5 2.45

Gibson 19 10 2.15

I have 3 words for anyone who hates this system: "Nolan Ryan 1987". I loved Nolan Ryan. I own an Astros jersey with his name. But Cy Young is the all-time leader with 511 wins. This system is explicitly designed to keep pitchers with 11 wins from winning his award.


edit: oh, how do I assign these awards? When the off-season starts, I go to the "Award Voting" screen and pick "Manual Award Selection". Then I click "Ask Computer" to get the AI choices.

After that, I change the MVP and Cy selection to my choosing and click "Finish Voting". BE CERTAIN. THIS CANNOT BE UNDONE.

I don't mess with the other awards. They're generally good enough and I'm too lazy. But the CY and MVP are too iconic to be lazy about.

Here's what it looks like. Also I set the Custom Award to "Houston Player of the Year". You know, local sportswriters do this for each team, so I like doing it, too. It guarantees that someone on my team gets recognized even if we have a terrible season.