In The Lab: A Look at the MVP Race

[ASFN Admin]

You must be registered for see images attach

There are any number of ways to look at the MVP award. It is why it is the single best award in sports. Some people look at it as who the best player is. Others look at which player meant the most to his team’s success. Obviously, the WAR revolution has taken the sport by storm. Some people live and die by WAR while other rail against it. In this edition we split the difference. There will not be one reference to a player’s WAR, but we should talk about what makes up WAR and why some people swear by it.

In this edition, we will look at six different players in the American League and their numbers coming into this week’s action. We will take two different looks. The first look are the traditional numbers every fan knows and care about. These are the numbers that have been governing the sport for 150 years. Therefore, a number of voters swear by those numbers and they use those numbers to sway their vote.

There are a number of issues with those numbers. First, it does not include a single mention of fielding, but we know that fielding plays a huge role in a player’s value. Secondly, we know that it is nearly impossible to compare the value that a shortstop or catcher brings to the table in comparison to a first baseman or designated hitter. There are issues like difficulty, complexity, but mostly positional scarcity. There are simply fewer great catchers than say first basemen.

Before I introduce the numbers we will actually use, we should take a look at the conventional ones. These numbers will point a certain way. Obviously, it includes seemingly sophisticated numbers like OBP and therefore OPS, but even when you include those numbers it has those nagging questions over fielding value and positional scarcity. Let’s see the first round of numbers.

AVG	OBP	SLG	HR	Runs	RBI	SB
Yordan Alvarez	.316	.431	.650	22	45	48	1
Yandy Diaz	.325	.399	.526	11	32	42	1
Ben Rice	.299	.392	.640	18	47	45	2
Bobby Witt Jr.	.280	.354	.456	9	33	27	23
Kevin McGonigle	.291	.394	.426	4	40	24	9
Nick Kurtz	.276	.431	.496	12	42	45	7

Bill James once described good numbers as having the quality of language. I would more call it art. The goal in all art is to capture the soul of something. Obviously, they do it in their own special and sometimes they miss the mark. I was never very good at art. I drew the sun in the corner of the page and had stick figures for people. Better artists add dimension. They add complexity. Their art is more lifelike and compelling. Good numbers do the same thing.

Runs and RBI are to statistics what stick figures are to art. They describe opportunity. Sure, we could further break them down and describe how often someone scores when given the opportunity or how proficiently they drive in runs. That would be more descriptive and would move the sun out from the corner of the page. Unfortunately, a hefty percentage of MVP award voters are still looking at stick figures. Some of the stick figures are quite impressive, but they are still stick figures.

Bill James developed a stat he called “total runs.” We have most of the components ourselves from his original formula (runs created + fielding runs + base running runs + positional adjustment) but we do not have all of his exact components. Instead, we will utilize baseball-reference.com’s version of all of those numbers. The first three are self-explanatory but the key difference is runs created. That is independent of the rest of the team, so luck or opportunity are muted. The positional adjustment (RPos) gives weight or takes weight away based on the relative difficulty and scarcity of the position. The upshot is that this is largely what WAR is based on, but instead of a number that is firewalled behind proprietary formulas, we are offering something more transparent.

RC	Rfield	Rbaser	RPos	TR
Bobby Witt Jr	45	7	5	4	61
Yordan Alvarez	70	-4	-1	-5	60
Kevin McGonigle	45	6	4	4	59
Nick Kurtz	55	2	0	-3	54
Ben Rice	55	-4	0	-4	47
Yandy Diaz	48	0	1	-5	44

It’s fair to ask what the difference is between this and WAR. In many ways this is like asking mom for her most guarded recipe. WAR will obviously include a ballpark normalization element that these numbers do not include. Obviously, the entire idea is to ask what would happen if a AAA regular were to be imported directly into that player’s spot. That player would almost certainly create runs and do the other things. Their finish here would not be zero.

So, WAR asks the question of how many runs better any particular player is than the AAA regular that would take that spot If it sounds complicated then that is because it is. Total runs simplifies that considerably. As you can see, the fact that Witt and McGonigle are good defensive shortstops matters. It helps close the gap to the point where they are virtually tied with Yordan.

All good numbers like this categorize players into like groups. Witt and McGonigle are similar to Yordan in value. Different numbers will project something similar but it will come out looking differently. Statistics should not determine winners on their own, but it should inform the voters enough to pick from a group of similar players. In other words, some could credibly argue for Witt and be on firm ground. I would argue for Yordan and be on firm ground. Others would go with McGonigle and be on firm ground. The numbers should hopefully prevent selecting someone that is obviously inferior. At any rate, total runs makes this look like a close race. What is your best argument for Yordan?

Continue reading...