1) Add up the number of runs that scored while the pitcher was pitching.
2) Subtract the runs that were scored by runners reaching base due to a fielder making a mistake.
3) Also subtract the runs that were scored by runners reaching base due to a pitcher making a fielding mistake. These runs are not earned for some reason.
4) Add back the runs that scored due to a fielder making a mistake in which he did not touch the ball. As long as a fielder really screws up but doesn't touch the ball, we're going to blame his mistake on the pitcher.
5) Subtract the runs that score after a fielding mistake (in which the fielder or pitcher touched the ball) that probably would have led to the end of the inning. The pitcher can give up 10 home runs in the same inning after such a mistake, but we will not consider those in this measurement of his performance. (Note: Do not subtract runs that score after a mistake that prevented what probably would have been the third out is the catcher interfering with a batter. For some reason the catcher keeping an inning alive via this mistake does not give statistical license to the pitcher to be as bad as he wants.)
6) Do not subtract runners that scored while other pitchers were pitching if those runners reached base (except in the event of a fielding mistake in which the fielder or pitcher touched the ball) while the pitcher whose ERA is being calculated was pitching. Even if the first pitcher only gives up 90 feet of base-running to an opposing player while subsequent pitchers give up 3 times as much base-running ground to that player, we still add that runner's entire 360 feet of base-running to the first pitcher.
7) Consider all defenses the exact same. Some will allow hundreds more balls to fall as hits than others will over the course of the year, but we're going to just blame (or credit) the pitchers for such variations.
8) Balls hit 410 feet that hit outfield walls and stay in play shall be treated the same as balls popped straight up but are not caught due to fielder confusion that does not involve the touching of the ball before it hits the ground. Do not subtract any runs that are scored by runners who reach base in either manner, except when either event takes place after a fielding mistake (in which the fielder or pitcher touched the ball) that prevented what probably would have been the 3rd out of the inning.
9) Do not subtract runs in which the runner reached first base due to a "wild pitch," but do subtract runs in which the runner reached first base due to a "passed ball." Some kid or sportswriter or accountant in his evening free-time will officially determine which pitches are wild or not.
10) Subtract runs that were scored by runners that scored after reaching base on a play in which a runner who had reached base by a fielding mistake (in which the fielder or pitcher touched the ball) is put out.
11) Subtract runs that were scored by runners that scored after reaching base on a play in whic h a runner who had reached base by a passed ball (but not wild pitch) is put out.
12) Subtract runs that were scored by runners who reached base without the aid of a fielding mistake (in which the fielder or pitcher touched the ball) but who at some point were able to move up one or more bases due to a fielding mistake (in which the fielder or pitcher touched the ball) or passed ball (but not wild pitch).
13) Now take the sum of those steps and multiply it by 9.
14) Then divide the product by the number of innings the pitcher has pitched during the time frame being measured.

Then you have a pitcher's ERA.

Fortunately this process is very simple and objective... unlike these crazy "sabermetrics" used by nerds who don't really watch the games or never played it.

Exactly. Kudos. Hope you were getting paid by some big rich international while you worked on all of those options.

I agree on Sabermetrics being completely stupid. WAR is the worst stat ever because it's all subjective! ERA is just the average number of earned runs surrendered in 9 innings so a complete game giving up 4 runs is an ERA of 4 and two complete games giving up 8 runs total is also 4.

Exactly...because there is nothing subjective when it comes to determining an earned run from an unearned run.

Rangers100 has become the new Joe, beating us down every day on the forums with his pet peeve. Yawn. Please wake me when it stops.

I don't even like to use ERA. The only way to truly decide how good a pitcher is are by wins and losses. No idea how King Felix won the CY last year, his W-L was terrible!

@pryor there is something over looked with ERA when it the charges the runner to the originating pitcher but does not take into account the pitcher that allowed the run to score. Similarly the pitcher getting a pass on runs scored after an error in certain cases has some logic to it but they should be accounted for somehow. With that said I am just dipping my toe into WAR, xFIP and such so I have no opinion yet on Saber.
At first blush it seems that the SABER analytic's are trying to take a look at more in-depth at ways to measure all things baseball.

I don't even like to use ERA. The only way to truly decide how good a pitcher is are by wins and losses. No idea how King Felix won the CY last year, his W-L was terrible!

LOL. Nicely done.

With that said I am just dipping my toe into WAR, xFIP and such so I have no opinion yet on Saber.
At first blush it seems that the SABER analytic's are trying to take a look at more in-depth at ways to measure all things baseball.

If you take both the ERAs and the xFIPs of a group of pitchers (with large enough sample sizes... say, 60 IP or more) in one year and then regress each against the ERAs of those pitchers in the following season (where the pitchers again threw 60 IP or more), you'll see that the xFIP numbers are far more closely correlated with the subsequent season ERA numbers than are the first year ERA numbers. In other words, xFIP is a far better predictor of future ERA than is current ERA.

Why?

Simple: because xFIP is a far better measurement of what the pitcher has already done with the things he can control.

xFIP just sounds weird, complicated, etc. It's foreign and unfamiliar to us. So people stick with ERA... because there is safety in that. No one will laugh or think the person is being a jerk or a nerd or whatever.

But if you can get beyond that and simply want to know which pitchers have performed the best over a given period, xFIP blows ERA away. xFIP is a statistician's statistic. ERA is just some silly, highly subjective gibberish created by sportswriters a century ago. You'd be better off just looking at RA than introducing the highly convoluted idea of "unearned" runs.

Rangers100 i am starting to get that. ERA is a hell of a lot easier to through out there in a TV broadcast though. Plus picture Rhads trying to explain it.

Yaaaaawn.

Rangers100 i am starting to get that. ERA is a hell of a lot easier to through out there in a TV broadcast though. Plus picture Rhads trying to explain it.

No doubt, mhilg.

I think with time we'll see FIP (and hopefully xFIP) incorporated into broadcasts, box scores, Jumbotron graphics, etc. But it will take awhile.

During a recent Yankees game I was watching on YES Network, David Cone spoke at length about WAR in discussing Curtis Granderson's season so far... and he did an excellent job of explaining it and concisely pointing out why it's a reliable and helpful stat. He had even had the broadcast team prepare an on-screen graphic that showed the AL WAR leaders (according to FanGraphs).

What has to happen first is for enough people to understand that pitchers - outside of a few extreme examples and even then only to a small extent - don't have significant control over the rate at which batted balls result in hits. This was only discovered by Voros McCracken 10 years ago, and it is so counter-intuitive and contrary to what we were all raised to believe about the game that many just refuse to believe it. But it's true and very well-documented at this point.

Once people understand and accept pitchers' uncontrollability of BABIP deviations from the MLB average (~30% with little deviation from year to year), then disregarding ERA is a natural next step (as it is very largely skewed by such random deviations). (Steve Slowinski at FanGraphs had a good post yesterday about explaining all this in simpler terms.)