Rabbit holes? In baseball?

I think what Caple was getting at in his article is that WAR should simply be another stat used to measure a player, not the end-all stat that some seem to think it is. There is no one solitary stat that is going to tell you everything you need to know about a player, it takes all of them collectively to gauge the specific player. This includes BA, RBI, W-L record...name a stat...to get a nice grasp on the value and greatness (or lack of greatness) of a player.

So BA and RBI are stats that shouldn't be taken seriously? Well, for the most part, those players with WAR numbers at 3+ have BA's above .285 (43 of the 73 at 3+ fWAR in 2012). And, the majority of them drove in over 75 runs (47 out of 73). .285 with 75+ RBI's on a yearly basis will net you a pretty nice contract, no?

If there was a single stat that described a player in full, I'm sure someone much smarter than any poster on here would have already come up with it. The fact that there are 2 measurements for WAR, and that each site's measurement differs so greatly on a few players, tells me that the WAR stat is fairly imperfect and shouldn't be relied on as heavily as some like to rely on it.

In a way, I think WAR reflects the increasing laziness of society. People want one stat to look at and determine the player's value, rather than looking at the entire stat sheet of a player. Who cares what I have to say anyway, I'm on the bus of people that aren't to be taken seriously because I still think a player's BA and RBI total matter.

Linear weights is based on mountains of historical evidence, determining the average number of runs scored in each situation, and thus, the relative value of each type of outcome in each situation. The only subjective decision in its construction is how far back to use historical data, as going back 30 years and going back 50 years to determine the weights of each outcome will result in slight differences, but at precisions that are not likely significant to the overall calculation anyways.

In other words, it's no more "subjective" than the calculation of batting average, as that is also inherently subjective (i.e., why is a walk not counted as an at-bat? who decided that???). At some point, you have to drill down far enough that something is axiomatic (basically, you recognize it as truth without need for further proof other than your own recognizance). Everyone has accepted at this point the definition of at-bat, and thus the role it plays in batting average. That doesn't necessarily make something like that a worthwhile metric for evaluating a player on the whole. WAR is a tool attempting to achieve the goal of evaluating a player as a whole (offense, defense, base running), while accepting some things as axiomatic (linear weights, for example, which gives rise to the calculation of wOBA... the math is really not something that is arguable at this point).

It's not that W-L, BA, and RBI "don't matter", as you put it. It's just that's it's widely accepted in the SABR community that these are not the best ways to measure performance, because they are not entirely dependent on the performance of the player in question. Pitcher's W-L record is dependent the performance of almost the entire team (offense and bullpen). Cliff Lee last year is a great example to show why W-L are misleading. Take King Felix's Cy Young season of a couple years ago as another example. Your recognition of greatness shouldn't be penalized by being surrounded with the crap he had around him in Seattle.

Batting Average is a misleading metric because it depends heavily on luck (see BABIP) as well not being a good metric of additional value provided by power. That's why OPS was invented, to try to capture the power aspect that BA misses out on. Based on the logic in this forum, doesn't that alone prove that BA is "imperfect"?

RBI's are a contextual based metric. In order to accumulate RBIs, you have to rely on the players ahead of you in the lineup to put you in situations where RBIs are attainable.

As was said, the point of Caple's article is that WAR isn't the highest pedestal of stat achievement... no one is arguing that it is. The fact that there are multiple calculations of WAR is a GOOD THING. If anything, we need more to achive a large enough sample size to accurately measure a player's total contributions.

WAR attempts to take these contexts, reliance on luck, and reliance on other players out of the calculation and show what the individual player contributed on his own.

When I said those stats don't matter, I was just quoting what someone said earlier. I think they matter, and to the casual fan, those are the easiest stats to understand. Those are the well-knows stats, the ones that are shown on TV when a pitcher takes the mound or a batter steps to the plate. You are right, BA is imperfect. All stats have some sort of imperfection to them, but that's the beauty of it. That's why all stats matter to a degree, whether luck is involved or not in the measurement. The performance of a team over the course of a season is filled with luck and dependant on each individual's performance. I understand the orgasms people have over WAR and all those other SABR community stats, but I like to try and keep in perspective that there is nothing wrong with looking at all stats when attempting to grade a player.

Adam, "Batting Average" is indeed objective. It is simply a number of hits per at-bats. We know how many hits there are, and we know how many at bats there are. There can be no argument over what the BA is, which is what makes it objective. HRs is objective, Hits is objective, Ks is objective, and so on.

But like all numbers, Batting Average and all the others can be USED subjectively, when we use them to define one hitter as "more valuable" than another overall. Or they can be used objectively, if we use it to say which batter is more likely to get a hit. But those stats have no opinion at the core. They are simple addition, subtraction, division, and so on.

Where subjectivity enters the equation is when intangible things like "value" and "wins" are said to be weighed, because then someone's opinion has to determine what creates value or wins. And subjective numbers are limited by the bias, rationality, and knowledge of the one who creates the formula to derive what it is that's trying to be determined.

Numbers aren't bad. There's always been a quest for defining numbers, but there is so much that they can't weigh, that they can only offer hints. When any one is used as if it is the be all and end all of analysis, it fails. When used as one tool among many, and when its limits are recognized (such as the fact that stats can only tell us a limited amount, and can give us info about what already happened, but can only give us clues as to what MIGHT be to come, and when used in looking forward can be completely wrong), it can help..

bb, I find myself disagreeing with the majority of your presumptive statements, such as: "Where subjectivity enters the equation is when intangible things like "value" and "wins" are said to be weighed, because then someone's opinion has to determine what creates value or wins. And subjective numbers are limited by the bias, rationality, and knowledge of the one who creates the formula to derive what it is that's trying to be determined."

This simply is not true. There isn't some nefarious puppet master sitting in a dark room with a mischievous grin on his face saying, "I don't like Michael Young so I'm going to give him a low WAR score. Oh! But there's Ian Kinsler! I like him so I'm going to give him a higher number."

It doesn't work that way. wRC+ and wOBA are based on formulas. xFIP and SIERA are based on formulas. UZR and RngR, defensive metrics, are predicated on the idea that "If player X_ can make that play, how many players relative to the same position can make the same play?" People that do this stuff for a living take their jobs seriously, so I'm not sure why they would award brownie points to players they like vs. players they don't. The concept is to be objective. Also, because we have BP, FanGraphs and BR, we have different outlets to quantify these numbers.

What I think we're driving at here is "What is a good stat?" vs. "What is a bad one?"

If I put two players head to head, which would you prefer:

(a) .325/.345/.475
(b) .270/.375/.475

Logically, you'd prefer the player with the higher on base percentage, no? That's why you'll never see me citing a player's worth based strictly on batting average. Stats like OPS don't do much for me, either, because it puts more weight on slugging percentage than OBP. One individual base (from a walk, for example) is more valuable than the difference between a double and a triple. wOBA and wRC+ takes these limits into account, which is why it's a more effective tool.

Eric, if my statements bother you by conflicting with your sense of the source and authority of formulaic stats, then that's great, because what I'm saying is simply observational and factual and isn't presumptive at all. The fact that you take them as such, indicates you're resisting the need to look more closely at those formulas you revere, and lashing out instead of thinking. But hopefully in time you'll open your eyes.

I'll try and explain one more time, and if you still don't grasp it, then so be it. I'd invite you to consider the info shared at this website for further explanation of the core concept: http://rationalwiki.org/wiki/Reasoning . Let me add that these are concepts I never understood intuitively, as we all take so much for granted in our everyday experience, and that I don't think others naturally understand either, but they can be easy to grasp once you look deeper.

When noting that WAR and other formulas for things like "value" and "wins" are subjective, that doesn't mean that some puppet master is juggling formulas to favor the player they like. Instead it's an observation that those formulas are based on someone's attempt to link a with b, which requires their own bias as to what is relevant and what isn't.

It doesn't matter which way you work the formula, either from value to stats, or stats to value; at some point, you will inevitably have to make cause-effect links and relative importance links in creating the formula. The more complex the formula, the more decisions being made, and the more biases that are embedded in the formula. That embedded bias is what makes results subjective and in essence opinions, rather than objective and facts.

WAR (or any other formulaic number) can make more sense to you as a determiner of value than others. But at the end of the day, if you depend exclusively on WAR (or any other) you're simply buying the opinion that whichever player's stats better satisfy a certain formula, that's who is most valuable. In that approach, one is ultimately buying an opinion packaged as a formula, and adopted it as their own.

^ Wow, incredible post.

Pretty sure everyone can tell whe it is me or when it is my secret admirer.

I did not post either of those comments. My my people sure seem to enjoy impersonating me.

Hey asshat, don't you have better things to do than obsess over a random poster on the Internet?

That wasnt me.. this is ridiculous.

pretty obvious who has the problem here.

I'm not even going to bother.

You two are both idiots.

If you're schizo, that makes four of us..

I was looking at the back of a T-205 card and noticed the stats for: Games, Batting, Fielding. (I think strides have been made.)

As long as Ted Williams ends up as one of the top 3 hitters I don't care what formula is used.

Lmfao. Cry more NTE. What's the matter? Can't handle the taste of your own medicine crybaby bitch?

I am truly sorry to say that bbtia is going the way of a blog site I visited along with Jumangi several years ago. bbtia is in the trash can.

It will be fine once Joey comes back and gets rid of NTE.

Or Romro.

Seriously, what is my fascination with Romro? I think I may have a secret crush on him.

Again with the GAY reference. Take another step and come on outta that closet.

Seriously, I know how good it feels to come out of the closet.