clock menu more-arrow no yes mobile

Filed under:

Analytics are good for player contracts, not bad. The Astros proved it.

If you buy something from an SB Nation link, Vox Media may earn a commission. See our ethics statement.

MLB: Houston Astros at Colorado Rockies Russell Lansford-USA TODAY Sports

Like clockwork this time of year, baseball’s Hot Stove season, baseball’s great debate over the fairness of player compensation slides back into the consciousness of media, agents, players, and fans. This year specifically, with numerous Free Agents seemingly held hostage on the open market while everybody waits for Bryce Harper and Manny Machado to sign, the argument rages because there isn’t a whole lot else to talk about in the baseball world.

But what is the argument, exactly? It’s hard to say, sometimes. Some appear to be hinting at owner collusion to depress player salaries. Some seem to have general outrage over MLB’s record profits that they perceive are not trickling down into the pockets of the guys on the field. Some seem to be making statements that sound terrible, but actually are just statements that can’t plausibly be connected with anything sinister.

And then there’s every traditionalist’s favorite whipping boy. Analytics.

What are analytics, exactly, other than something that Charles Barkley is sure that the rest of the world disdains, and are only used by virgin nerds who can’t bench forty pounds? Merriam-Webster defines analytics as, “the method of logical analysis”.

OK.

So doesn’t that make...you know...using batting average and ERA to determine if a player should get more or less money...analytics?

I digress slightly.

What do analytics have to do with the systematic robbery of MLB’s veterans? Everything, according Giants third baseman Evan Longoria.

So here is Longoria’s theory in a nutshell:

“It seems every day now someone is making up a new analytical tool to devalue players, especially free agents.”

Mkay.

Set aside the fact that this is coming from a guy who signed two of the team-friendliest contracts for a superstar player ever signed (10Y $135M, back in 2013, and before then 6Y, $17.5m), and so he has no business accusing other people of “devaluing players” monetarily when he’s setting that kind of example.

First, there’s a more obvious reason for owners to want to depress contracts that doesn’t require inventing player-sabotaging analytics.

Let’s run with his premise that owners are systematically trying to pocket more of their profits by inventing statistics solely for the purpose of depressing player contracts and see where it leads. (warning: some numbers stuff ahead, but it’s worth reading and understanding)

To do that, we must first ignore the MLB salary cap. MLB doesn’t have a salary cap, you say! Well, yeah it does, even if it’s not a hard cap. MLB’s luxury tax threshold was collectively bargained by teams and by the MLB Players’ Association thus:

When a team goes over a certain payroll number (in the ballpark of $200 million, escalating a bit every year), they must pay an additional 20% to MLB for every dollar they go over. If they’re a first-time “offender”. If they’ve done it before, 30%, and then 50% for third and higher.

In addition to that, there is an additional 12% tax on payroll that goes more than $20M over the luxury tax, and if a team goes more than $40M, there’s a whopping forty two point five percent surcharge.

Also, that last group going over $40M also have their highest draft pick dropped ten spots, have harsher penalties for signing a player who has declined a qualifying offer, and if they lose a qualifying offer player to another club, get a compensation draft pick after the fourth round instead of the first!

In 2018, the Nationals were slightly over the luxury tax threshold and had to pay about $2.5 million in penalties (the Red Sox had to pay $12M, FYI), the first season in which the collective bargaining agreement took full effect. So the Nationals, as first time offenders, would see a bigger financial hit if they go over again in 2019.

They currently sit with a payroll of $151M, a figure that will increase to around $160M as they work their way through arbitration. Let’s pretend that the Nationals are going to spend a lot more money, to relieve the pressure on the poor cash-strapped veterans who all of the other clubs have stolen money from by the nefarious means of analytics.

In a fit of excitement, they shell out $70 million to Harper and Machado and say, “Hey, we’ll just make this work somehow,” and then they blow another $20M on Dallas Keuchel.

That would be a World Series favorite, no doubt, and satisfy Longoria’s demand to damn all consequences and just sign the best players. But by being more than $40M over the cap again, the Nationals would have to pay something like $29 million in tax dollars to MLB, and also lose the value of all of those draft picks and stuff too. Let’s call the total bill in hard cash and cash value of picks to be $35 million.

What that means is that the Nationals didn’t sign Harper, Machado, and Keuchel for only $90 million. The actual cost to the Nationals is $125 million for those three players, clearly more than the players are actually worth compared to the rest of the market, even as good as they are. If I am a fan, I am deeply concerned about the irresponsibility of the decision makers in the front office, and I worry that the success will be short-lived, leaving me and mine suffering the consequences of a losing decade due to fiscal irresponsibility. Sound familiar, Astros fans?

This is what is actually holding free agent contracts in check, not some nebulous concept of “analytics”. The example above is an extreme one, but if you take the premise and don’t apply it to just MLB’s most expensive superstar free agents, but spread the principle of spending carefully because of the tax penalties among all free agents, you start to see how the market is responding to the terms of the CBA in a reasonable manner.

In fact, the ire is misplaced towards the owners, and perhaps should be directed at the superstars who are demanding such a large percentage of the pie for themselves that they are depressing the earning ability of all of the other free agents.

Owners and GMs know that to land a superstar, you have to pay far, far, far more for that player, and that the player can often name his price when several teams are bidding. To be able to afford that superstar, a team must save the money elsewhere or face the very real financial impacts described above. This is a direct impact of the CBA luxury tax agreement and of the current going rate of MLB superstars. Take that up with the Player’s union, Longoria.

Because the luxury tax applies only to player payroll, the only way a team owner can sign a superstar is to save money by spending frugally on the players who aren’t superstars...or to not sign average to below-average players at all.

The only thing new-age analytics has done is more clearly define which players are worth the extra money and which ones are inefficient use of the limited cash that the CBA allows teams without facing stiff penalties.

Average to below-average major leaguers are valuable, but only up to a certain price. That price point, in the current market, is probably close to the arbitration-negotiated values those players are getting in their 2nd or 3rd year of arbitration. Otherwise, more players would be non-tendered during those off-seasons than there are.

We all know that Free Agents are paid better than arbitration-eligible players. But with a limited pool of funds, the jump in cost between today’s arbitration-eligible average player and yesteryear’s average free agent is a wide gulf.

Is that a bad thing? Was it good that in past decades, mediocrity was rewarded handsomely while younger, cheaper, and often better players were buried on the depth chart?

The aggressive Luxury Tax penalty system is also helping players

This is not obvious at a casual glance. The reason behind the stiff Luxury Tax penalty system, and likely why the Players’ Association agreed to it in the first place, is for competitive balance across the league.

As it is, media pundits complain vociferously about the number of teams they claim are tanking for better draft picks. But how much worse would it be if teams have a better understanding about whether or not they’ll be in contention via analytics, AND a lack of stiff penalties allowed unchecked spending by the “haves”.

The fact is, there are only a finite number of roster spots in large-market cities like New York, Chicago, and Los Angeles. Even if those teams satiated a full roster’s worth of players to the tune of a $400 million payroll, improved understanding of probabilities would make it less likely for the other twenty teams in MLB to spend big in Free Agency. They will know that their only quasi-viable (but not very viable) path to victory lies through drafting and player development, after which their hard-earned victory will turn into losing their best players via Free Agency to a club that can double whatever they are able to offer and stay profitable.

The end result? A top-heavy league of perpetual high-spending winners in cities where most baseball fans will be concentrated. Because in those twenty other cities, fans tired of seeing their favorite players move to the East or West coast and tired of long years of losing will turn to other sports.

That in itself will lead to further depression of player payrolls as MLB starts to struggle for television ratings in ninety percent of the country.

And that brings us to the main point that Longoria and others who think like them are completely missing:

There are a number of players today who are handsomely rewarded in arbitration and Free Agency because of analytics, not in spite of them.

One needs look no further than arguably the most analytic club in baseball - the Houston Astros.

By the end of 2013, Collin McHugh had a major league ERA of 8.94 after having pitched for two different clubs. In early 2014 he wrote a blog post titled “Good, but not good enough,” In which he philosophically seemed to realize that the prototypical Major League dream might be beyond him, and in fact is beyond all but a small percentage of people who have ever played.

That was two months after the Astros claimed him on Waivers. One month later, he wrote about his report date with the Astros, the day on which he was told, “We’re really glad to have you here, son. You belong.”

The Astros weren’t just making a down-on-his-luck depth player feel good just to be nice, either. They meant it. They told him immediately to ditch his sinker entirely and throw more four-seam fastballs up in the zone. Up in the zone? Yeah, they said. Because if it comes in at the level of your curve ball, batters won’t know which pitch is which until it’s too late. And you have a very good curveball. Throw that more, too.

The result was 25 major league starts in 2014 with a 2.73 ERA and a strikeout rate of more than a batter per inning, the 15th-lowest ERA in the major league among starters.

How did that happen? Analytics. The Astros used data measuring curveball spin rate and data measuring batter reaction time to determine that if McHugh threw his fastball and his curveball on the same plane, a batter is physically unable to react in time to adjust to one pitch or the other. He has to get lucky to hit the ball, or McHugh would have to make a mistake.

The sinker, the Astros determined, was so different from either pitch that batters easily picked it out and hammered it. It wasn’t that good of a pitch anyway, with low velocity and uninspiring break. Which they measured. And compared to other players’ pitches. And the outcomes of those pitches. Going back years. To determine that McHugh’s was ineffective. Analytics.

MLB: Houston Astros at Texas Rangers Andrew Dieb-USA TODAY Sports

Fast forward to today. In 2018, McHugh was one of the best relievers in baseball, pushed into that role only due to the hilarious depth the club had in the rotation. In 2019, he will be the club’s third starter.

Courtesy: baseballsavant.com

In 2019, McHugh will be making $5.8 million, after making $5 million in 2018 and $3.5 million in 2017.

A pitcher who had been instructed to throw his sinker a lot by two other clubs and was on the literal edge of being out of professional baseball forever, has since earned over $16 million in the five seasons since being cut by his previous club and then salvaged by Houston.

Next season, McHugh will be a Free Agent. With his track record as both a starter and a reliever with the Astros, he will probably be able to command a contract eclipsing the one Lance Lynn just signed with the Rangers, 3 years and $30 million. Thus, his potential earnings between the ages of 27 and 35 will be around fifty million dollars.

Because of Analytics.

In 2012, Will Harris was waived by the Rockies after only 17 innings in the majors, in which he posted an 8.15 ERA. He was claimed and then waived by the A’s, and then claimed by the Diamondbacks, who waived him after posting a 4.34 ERA in 2014 over 29 innings. Even publicly-available analytics showed that Harris was better than his ERA, but he had already been waived by three clubs and could have been on his way out of baseball.

When the Astros claimed Harris, they didn’t even need to change anything, because they understood the statistical implications of small sample sizes, high BABIP, and how WHIP told a better story about pitcher ability than ERA does. Harris posted a 1.90 ERA in 2015 with the Astros, and has since been one of the most consistently successful and reliable relievers in baseball. He has since earned over $10 million with the Astros, and will hit a market next season, where he can expect to command the type of money earned by Adam Ottavino or Kelvin Herrera this offseason, around $8 or $9 million per year over several years.

Because of Analytics.

Analytics don’t just help the earning power of diamonds in the rough. Future Hall of Famer Justin Verlander looked to be nearing the end of his career in 2017 before being traded to the Astros. He held an ERA of 3.82, the highest it had been since 2014 and the third highest ERA of his long career. His FIP and xFIP indicated that worse was to come.

Famously, as soon as the Astros had him under their wing, they sat him down and pointed out measurable evidence using Pitch F/X and TrackMan that over the past season with the Tigers, his ball release point had slightly changed, which was affecting the behavior of his pitches. They helped him correct that, and he posted a 1.06 ERA en route to helping the Astros win their first-ever World Series championship. He followed that season up, at age 35, with arguably the best season of his career, posting 2.52 ERA with a strikeout rate that was more than three K’s per nine inning higher than his career average, all while lowering his walk rate to more than 1 walk per nine lower than his career average.

MLB: ALCS-Boston Red Sox at Houston Astros Troy Taormina-USA TODAY Sports

When Justin Verlander hits free agency at age 36, he will still be able to command a multi-year contract that will likely be at or in excess of $25 million per year.

Because of analytics.

Charlie Morton was an average and injury-prone major league starter who didn’t boast any special ability to strike out batters prior to coming to the Astros. The Astros gave Morton a 2-year contract worth $14 million despite pitching only 17 innings in 2016 due to an injury, and were called crazy by media pundits. That signing could never have happened in the pre-analytics era.

Like McHugh, the Astros told the man known as “Ground Chuck” for his groundball-inducing sinker to....stop throwing the sinker so much. Why? Because he could be a “super McHugh,” armed with a backbreaking curveball that paired not with a 92 mph fastball like McHugh, but a 98 mph fastball. He could, they said, be deadly to batters.

Morton then went on for the next two season to post two of the three lowest ERAs of his career at ages 33 and 34, and he struck out nearly four batters per nine innings more than he had ever done in his career. He finished up 2017 by closing out the World Series with four dominant innings that tore the soul out of an over-matched Dodgers lineup.

MLB: World Series-Houston Astros at Los Angeles Dodgers Robert Hanashiro-USA TODAY Sports

At age 35, Morton signed a $30 million contract, something that would have been impossible had he continued following the career trajectory set out for him by the Pirates.

Because of analytics.

So no, Longoria, analytics are not depressing player contract value.

Analytics are assisting players who would otherwise be unnoticed and enabling them to be rewarded with contracts they never could have realized otherwise. McHugh was close to being forced to hang up his cleats and try to make a living using whatever he studied at Berry College. Harris might have bounced around Triple A for a while before giving it up and doing something with the “General Studies” degree he earned at LSU.

It’s very hard to argue that analytics are depressing player contracts when analytics are directly responsible for setting up many players for life who would have been completely ignored or overlooked in the so-called glory days of baseball, when we hallowed hard-nosed guys that roped singles galore but provided little value otherwise, or ignored pitchers who threw “only” 90 miles per hour in favor of harder throwers who couldn’t throw strikes.

But analytics will continue to be the whipping boy for people who refuse to acknowledge reality.

It’s a classic fallacy to blame something you don’t understand when confronted with a situation you don’t like. And this is exactly the case here.

The market situation in MLB has changed with the advent of the newest collective bargaining agreement, in the interest of competitiveness and a better fan experience in smaller market teams.

Despite this newly-constricted market environment, analytics has allowed diamonds in the rough like McHugh to emerge, and allowed veterans like Morton and Verlander to extend their careers and expand their earning power.

Yes, analytics have also identified players as “average” by finding deficiencies that might not have been so readily noticeable back in the day. For example, C.J. Cron, who was released by the Rays after 2018, despite hitting 30 home runs that season. But “analytics” showed that the power came with low on-base percentage and terrible defense that depressed Cron’s overall value, to the point where the Rays didn’t want to pay the likely $5 million required to keep him. They felt they had better options for their club and for their fans...for cheaper. Some of that money saved went to Charlie Morton, who the Rays felt could contribute more to the club’s win total. Analytics say that they are probably correct.

Do the Rays have payroll space to pay Cron anyway? Probably. But the Rays believe that Yandy Diaz and Ji-Man Choi can do a better job for a tenth of the cost. Cheaper and better.

The Rays could be wrong, but in days past, clubs might have blindly re-upped because of those 30 bombs. Back then, all those other considerations would have been ignored because OH MY GOSH HOME RUNS. In that market, Cron is rewarded for possibly being worse than the other two guys mentioned, whose own earning powers in turn suffered for it. Is that more fair than the way it is now?

So yes, from an individual player earnings standpoint, analytics can be a savior or a devil. But who deserves that money anyway? The player who is actually good but improperly-used and overlooked? Or the player that is flashier, but doesn’t make the team better as much?

The market, and analytics, have sorted that out together.

Is it really a bad thing that the market and analytics have resulted in a more balanced and better fan experience?

Are we really outraged because mediocrity is no longer being rewarded with out-sized contracts, while more talented or under-utilized players were being marginalized?

Sorry, but the narrative is false. The “human interest” benefits of analytics helping players reach their best potential and therefore maximizing their earnings far outweighs the impact that analytics have on the earnings of those who are mediocre.