WAR

Without him in the lineup the As were 12-22 for a .352 winning percentage. Over a full season that projects to 62 wins.
His WAR according to Baseball America was 3.4. So replacing the proverbial replacement player with him should improve their record to 66 wins. I rounded up to be generous.

With him in the lineup the As were 82-46 for a .641 winning percentage which projects to 103 wins. So tossing him aside for the proverbial replacement player that drops to 99. So that WAR thing was only 33 wins off.



Give me real stats over imaginary ones to evaluate players any day of the week.

I suppose this argument makes sense if little things like correlations between cause & effect don't mean much to you...

WAR...

What is it good for?

I suppose this argument makes sense if little things like correlations between cause & effect don't mean much to you...

Cause and effect? You mean like adding his power to the middle of the lineup which made a world of difference from the first time I saw Oakland play and the last time i saw them play.
Cause = Cespedes added to lineup
Effect = Team turns around completely.

Anyone who watched Oakland from beginning to end knows Cespedes meant a great deal more than just an extra 3.4 wins to Oakland. Without him they were a last place team. After his arrival the climbed to top of division.

A similar scenario was Mike Trout. That lineup looked totally different once he arrived.

But I understand the concept. They are in theory trying to compare what a team would do with a given player as opposed to an average replacement from AAA. First flaw is I think it overcredits what the average AAA player can do at the major league level. Second is the saber guys have gone all the way back and assigned WARs to players from the 1800s and early 1900s when minor leagues did not exist and that hypothetical replacement player was some Joe Shmoe from off the streets. It makes for an interesting discussion topic but gets used way too much to evaluate a player's season or contribution to team.

Wow, without even arguing the merits (or lack of) in WAR, you are attributing the entire difference in record based on one player in your statement. If only life were that simple.

Forgive me for cluttering the WAR discussion with actual facts.

Actually I am just giving one example. But if it means anything it should hold up for any and all players. Are you going to seriously tell me given the change in the Oakland team after Cespedes arrived he was only worth an extra 3.4 wins? And are you going to attempt to do that with a straight face?

I could give more. I already mentioned Steve Trout. Anyone who wants to bother with actual research will see that rarely does WAR pan out when you compare what happens when a MLB player goes down to injury and has to be replaced with an actual replacement from minors instead of the hypothetical replacement player.

Look at Mike Napoli. His BA was down but he hit 24 HRs last year but WAR is only 2. So I am to believe thoses 24 HRs were only worth 2 wins and Rangers could have replaced him with their AAA catcher and only won 2 fewer games.

I would bet the average person who quotes WAR numbers when comparing players could not even tell you how it is calculated.

Face it the thing is not reliable. Those who blindly follow the WAR propaganda will do so no matter what I say.

Actually, Steve Trout injured himself falling off a stationary bike, see the attached link.


http://bleacherreport.com/articles/7636 ... ry/page/10

But I do agree with you, Valen, one player can and does make a difference, if it is the right player - one that can elevate the play of his teammates by his mere presence and leadsership by example.

Great insight Risden. No matter how complex your formula gets few if any can accurately measure the impact any given player has on the other players in the lineup. As a Ranger fan I watched the impact Cespedes had on that Oakland team. Before he arrived I almost considered a game against the As to be a vacation day for our pitchers. Time to pad the stats. But not long after he arrived and as the season progressed my thoughts were more along the lines of we better have the bats warmed up.

For those who consider WAR to be the end all for measuring a player's contribution to winning consider this:

only four of the top 12 American League players in terms of WAR and just six of the top 20 even appeared in the postseason. In other words, WAR said they were winners but they failed to contribute to their team winning.

Consider this from baseball-reference.com

Sports Reference sets replacement level at a .320 winning percentage for recent seasons. This means that we expect a team of replacement players to have a .320 win-loss percentage or a 52-110 record.

Now consider World Series champion San Francisco led the NL with a 28.9 WAR. So that means SF should have won 81 games and that would get them a world championship. You should not need a degree from MIT to see something is wrong with that.

The Giants with possibly the best pitching staff in the NL only had a pitching WAR of 5.5, 13th in the NL, barely ahead of Houston who was at 5.4. So the Giants almost had the worst pitching staff in the NL and probably could have swapped out their staff with their AAA affiliate and still won it all. Any stat that tries to tell me those 2 teams pitching staffs were about equal is fundamentally flawed. It makes for interesting conversational material but seriously Houston and SF pitching staffs equal???

Now if I upset anyone too much don't worry. Pitchers and catchers have reported so very soon I will not be nearly as bored or interested in yanking anyone's chain.

I'm in the sabermetric is crap camp. WAR is relative-so replacing a great player off the Giants has very little impact relative to taking a great player off a bad team.

I agree with your point scumby.

For me the formula breaks down because replacement player is too murky. When I first read about this stat the narrative was relative to whether a team was better off going with a given player at whatever salary verses giving your best player in the minors that slot. The argument essentially went Joe Minorleager could give you these numbers so any player on your roster who will give you less is performing below replacement value and is just standing in the way of a young guy getting a shot. It then morphed in to this formula where every player is measured against this hypothetical replacement player.

But what is that replacement player really? Is it as Sports Reference states the level of player that would produce a .320 winning percentage? Is it the average AAA player? Is it the best AAA player at a given position. What if some other team owns that player. If so that is not your replacement player. Case in point, the Rangers. If Elvis gets injured or traded the replacement player for him is Profar, the #1 prospect in baseball. If they lose their catcher the best replacement they have in the minors are guys i suspect have trouble ordering bats because no bat company is going to want to ruin their reputation by have that guys picture took swinging their bat.

I know there are people who claim they can take a minor leaguer's numbers and convert those to major league equivalencies and tell me how that guy would perform if promoted right now. But all you have to do is match up some of those past predictions to how those players actually performed upon reaching the show to know those guys are full of it. And if you cannot know how that replacement player from the minors will perform how can you truly know how much better any given player is than that player?