Author Archive

Substitution Rules

In September of 2013, Billy Hamilton made his MLB debut. As a pinch runner. It’s wonderful to see Billy Hamilton be a pinch runner. A few days before his 20th birthday, Tim Raines made his MLB debut as a pinch runner, and all six games in his September callup was as a pinch runner. He played 15 games the next season, 7 of which was as a pinch runner. (The season after that, 1981, is when he established his star.)

I see nothing but positive about this kind of development. The only reason we see it in September is because of the loosening of the roster rules. Now, rather than expanding the rosters during the season, or declaring a 15- or 20-man roster on a game-by-game basis, what would happen if we changed the substitution rules?

Say you bring in Hamilton to pinch-run for the catcher. At the end of the half-inning, the manager currently has two choices: (a) choose whether to bring in a new catcher and knock Hamilton out of the game or (b) double-switch Hamilton into the game, with a new catcher. In either case, the original catcher is knocked out of the game. But, what if we add a third option: (c) allow the manager to keep the original catcher while knocking Hamilton out of the game. The substitution still knocks one player out of the game, but now the manager gets to decide which of the players involved gets the boot.

And you can extend that to pinch hitting as well. And if you do that, you can also do away with the DH. You can now bring in a pinch hitter for your starting pitcher in the second inning, without knocking out the pitcher (but you do lose the PH): You simply let the PH hit, and then that guy is out of the game, while the pitcher remains. And the same applies for a defensive substitution: at the end of the half-inning, you decide which of the two players is knocked out of the game.

You still have to worry about your bench and when to allow your pitcher to bat or not. But, you now give the manager a bit more flexibility. And if he wants to have a speedster on the roster, without thinking he’s burning through two roster spots (i.e., knocking out both the original catcher and Hamilton), he’s only burning through one.

What we have here is a rule that:
a. has no roster impact (you always lose a player)
b. is unobtrusive (no one is really going to notice anything)
c. allows the DH to disappear from the rulebook
d. gives the manager great flexibility

Downside? You tell me.

UPDATE: Based on one or two comments, I don’t think I was clear enough: at the end of the half inning, the manager chooses which player he loses for the game. In the above situation, if he goes for option c, he’d knock Hamilton out for the game, and the original catcher is allowed to stay. The same applies for a fielding sub: Hamilton could have come in as a defensive sub in the 8th inning in CF. At the end of the half-inning, the manager decides whether to knock Hamilton out of the game, or knock the guy he had replaced out of the game. There is NO “re-entry”. Someone will get knocked out of the game, just like it is today.

Scouting Report 2012 – By the Fans, For the Fans

Last callout!

The results of this project have appeared on the pages of Fangraphs for the past few years. Please take a few minutes to follow the instructions and fill out the ballot of your favorite team.

This is the tenth annual ballot, and its success is entirely dependent on your participation!

RE24 –

This post originally appeared on

One of my favorite stats is RE24, which goes by other names, like “value added” or “value added by the 24 base-out states”.

The basic idea is that you are interested in the 24 base-out states, and the outcome of the performance in each of the particular states. A HR with bases empty has a different impact than a HR with men on base. A strikeout with a runner on 3B with less than two outs is hugely impactful, while with no one on base, it is no different than any other out.

To the extent that you think a player should be recognized for that outcome in that context, then RE24 gives you exactly that.

Read the rest of this entry »

Community Playing Time Forecasts

It’s time for the annual community playing time survey.

Find the team you follow, and take a couple of minutes to fill it out.

Forecasting Mailbag

There seems to be some confusion as to what forecasting systems do and don’t do, and how to interpret the results.

Post your questions in the comments section, and I’ll answer them.

Evaluating Pitchers as a Concept: Average, Replacement Level, or Just Totals?

Don’t read further until you’ve already participated in this thought-exercise Cy Young poll.


The intent of this poll was to determine how the Fangraphs readers evaluated pitchers with different amounts of innings pitched. Basically, how does a reader balance better rate stats with lesser playing time stats?

The Pitcher X that all other pitchers were being compared to had a 13-4 W/L record, 2.00 ERA, and 153 IP. The league average pitcher has a 4.00 ERA. (You can presume that all the other missing stats would be consistent with that kind of W/L and ERA record.)

Read the rest of this entry »

To Replacement-Level or Not?

NOTE: If you haven’t seen the poll, then click that FIRST, then come back here to read more.


This is a the final part of a three-parter (for today anyway).

One thing that I wanted people to consider is that adding to one guy is like subtracting to another guy.

Say we look at our two players:
Player X: 105 runs created in 105 games
Player CD: 125 runs created in 162 games

The typical replacement-level process is to start with this guy:
Baseline: 0.35 runs per game

And we subtract that from each player.

So, Player X goes
from 105 runs created in 105 games
to 105 – 105*.35 = 68 runs created above Baseline

And Player CD goes
from 125 runs created in 162 games
to 125 – 162*.35 = 68 runs created above Baseline

Therefore, in terms of runs above replacement, both are at 68 runs.

But, what if instead of subtracting as I’m doing here, I simply ADD 0.35 runs per MISSING game.

Now we have this:
So, Player X goes
from 105 runs created in 105 games
to 105 + (162-105)*.35 = 125 runs created WITH baseline

And Player CD, having played all 162 games, remains at: 125 runs created

See? In both cases, we get the exact same answer.

When it comes to MVP talk, I presume a fair number of readers can’t fathom giving runs to a player for missing a game. That those 57 missing games should get zero runs, and therefore, the 105 runs in 105 games must remain identical in value to 105 runs in 162 games.

And I also think that those who support replacement level don’t realize that they are giving credit for the missing games, that they are in effect adding 20 runs to our Player X here.

In the end, it all comes down to an equivalency. You have someone with 105 runs created in 105 games. Is that better or worse, for MVP talk, than someone who created 106 runs in 162 games? How about 109 runs? 112?

The average Fangraphs reader made that decision: the average is 125 runs created in 162 games is equivalent to 105 runs created in 105 games. And so, the average Fangraphs reader supports adding 0.35 runs per game, for every missing game.

How Much is Playing a Game Worth?

NOTE: If you haven’t seen the poll, then click that FIRST, then come back here to read more.


I asked this question in a poll:

Player X created 105 runs in 105 games. AFTER which player would you slot him in, wrt MVP?

And I gave a list of Player A (150 runs) to Player F (100 runs). But in all cases, those guys played 162 games.

A straight arrow reader deduced the true intent of the question:

So the question is how you handle 57 games of production from somebody else.
a. Do you ignore it completely and just judge the guy on his 105?
b. Do you assume replacement level production?
c. Average?

If you simply give this player zero credit for the missing 57 games, then you would slot this player who created 105 runs in 105 games in between Player E (110 runs in 162 games) and Player F (100 runs in 162 games).

But perhaps you want to go the other way, and figure that the missing 57 games will be picked up by an average player, and so, you credit Player X with those runs (28.5 runs in this case), and so his 105 runs in 105 games is EQUIVALENT to created 133.5 runs in 162 games. Therefore, you’d slot him between Player B (140 runs in 162 games) and Player C (130 runs in 162 games).

Perhaps you think both those options are unfair:
(i) it’s unfair to presume an average player would pick up the slack
(ii) it’s unfair to presume that those 57 games would be a complete black hole

So, perhaps you decide that those 57 games need to get some runs credited to our Player X. Since the average player would create 28.5 runs, and a black hole player (i.e., pitcher as batter basically) would create 0 runs, then maybe something in-between, say 14 runs is what you should count.

Our Player X, with 105 runs in 105 games would be equivalent to a player with 119 runs in 162 games. And so, you would slot that player between Player D (120 runs in 162 games) and Player E (110 runs in 162 games).

The consensus pick was between Player C (130 runs in 162 games) and Player D (120 runs in 162 games). Therefore, Player X (105 runs in 105 games) would be equivalent to a player with 125 runs in 162 games.

Mathematically, you’d write this as:

105 – 105x = 125 – 162x

Rearranging the terms:
20 = 57 x

Solving for x gives us x = 0.35

Therefore, we give our Player X a rate of 0.35 runs per missing game.

Since the average rate (as noted in the poll) was 81 runs in 162 games, or 0.50 runs per game, then our “replacement level” is 0.35 runs per game, or -0.15 runs relative to average.

And -0.15 runs per game times 162 games is 24 runs per season. That’s where you readers have established the replacement level. And, as luck would have it, that’s pretty much exactly where saberists like to set the replacement level.

How you chose your answer is exactly how you handle replacement level for MVP discussions. For those who slotted him between the 110 and 100 runs created player, then you don’t believe that you should use replacement level for MVP discussions.

Game Score (and Crowd-Sourcing)

Bill James created a metric called Game Score that looks at a pitcher’s standard pitching line, to come up with an overall score, centered at 50, with most scores in the 0 to 100 range. I am trying to take that concept, deconstruct Game Score, and reconstruct it by forcing it to follow some set rules.

The scale of Game Score will mimic win percentage. So, a Game Score of 50 means you will win 50% of the time. A Game Score of 70 means you will win 70% of the time, and so on.

We’re not going to be total maniacs about it, and force a Game Score to stop at 99 or 1. Think of the relationship between Game Score and win percentage more as useful guidelines, rather than a hard constraint. This is especially because the relationship between wins and runs is not linear, and so, it’s going to be impossible to create a linear metric that will work at the extremes.

We also know that ten marginal runs equals one marginal win. This means that one marginal run equals 0.10 marginal wins. In terms of the Game Score scale, one marginal run is 10 Game Score points. Keep this in mind as you read the various versions of Game Score.

In addition, we’re going to set the starting point of Game Score to 40, rather than following the Bill James lead of starting at 50. The idea here is to think in terms of replacement level, and if you pitch to one batter and are out of the game, we’d hardly call that an “average” game. Indeed, I would even consider starting the Game Score at 35 or even 30. For the moment, we’ll start at 40, and let’s see where this takes us.

Read the rest of this entry »

Community Forecast – Playing Time 2011

Help me – help everyone – help you.