Putting Nick Markakis’ season in context

Unlike some hitters, Nick Markakis did not suffer through a sophomore slump. He also didn’t suffer through a junior slump. On the contrary, those were the years he rampaged through the American League, putting up a 4.1 fWAR season in 2007 and then a 6.0 fWAR season in 2008. In that two-season stretch, he was the 20th-most valuable position player in baseball.

After that, though, Markakis settled in as an average regular, but not much more. Over the next six seasons, he averaged 1.8 fWAR, and consistently fell pretty close to that mark. The furthest deviation from this average was his very uncharacteristic down year in 2013 (87 wRC+, 0.5 fWAR in 700 PAs); every other year, he came within 0.6 fWAR or fewer of the 1.8 fWAR average. In aggregate, he was only the 113th-most-valuable position player between 2009 and 2014, despite having the fifth-most PAs of any player in that span. (Even Hunter Pence, one of the five players with more PAs than Markakis over this span with the lowest cumulative fWAR, had nearly twice as much, 20.8, as Markakis’ 10.7 in this stretch.)

By the time Markakis signed with the Braves, he was into his 30s, and appeared to be experiencing the expected age-related decline. He matched his most recent pre-Braves wRC+ in 2015 (106), and then saw two consecutive declines (98, 95). His fWAR for his final Baltimore season was 2.3; he followed that up with totals of 1.4, 1.2, and 0.7. Much like how he looked pretty consistent in his late 20s, Markakis looked to be consistently declining in his 30s, playing out the string on a pretty nice major league career that saw him accumulate more than 25 fWAR.

All of this can be easily summed up visually.

Everything in that chart is consistent with general player aging and progression, including the early peak and the decline in the 30s. The glaring inconsistency, of course, is the uptick at the right-hand side, indicating that Markakis is on track for his second-best season at the age of 34.

The thing is, this is actually happening. While there’s no guarantee that the second half of Nick Markakis’ season will resemble the first half, his renaissance is not a trick of the light. His walk rate hasn’t slipped, his strikeout rate has taken an even further dip, his exit velocity is up, his launch angle has been inconsistent but is kind of up, his xwOBA is a ridiculous .390, and 45 percent of his contact is at 95 mph or more, which is also way up. Even his sprint speed is up, somehow: he’s gained 0.4 feet per second on the 2017 version of himself, and 0.8 feet per second on the 2015-2016 version.

Even his defense is actually happening. While it’s difficult to disentangle the effects of positioning from the readings of defensive metrics, he is on pace to post his highest UZR since he was a spry 24-year-old having a career year. His DRS puts him on pace to have his second-highest mark since that career-high mark in 2008. After posting fairly dreadful -5 Outs Above Average (OAA) marks in each of the last two years, he is currently positive with 1, and his catch percentage added is a neutral zero instead of the slight negatives he’s posted. He’s not just hitting better — he’s moving better, too, and it’s making a difference. Juan Ponce de Leon should give him a call.

All of this raised two questions for me. First, what is even the likelihood of this happening? And, second, what does this mean for Markakis going forward? After all, he’s going to be 35 next year. 35 is commonly thought to be a threshold for even more rapidly-accelerating age-related decline.

How Common is A Post-30 Resurgence?

Luckily, this question is fairly easy to answer. For consistency’s sake, I used the following criteria:

• 2002-2018 (why? because 2002 is the start of the UZR/DRS era, mostly, but using that range also has many other advantages)
• Minimum PAs in any season = 200
• Nick Markakis had three consecutive seasons in his 30s (his only seasons in his 30s) below 2.0 fWAR/600. A review of the data indicates that there are many players who have an occasional down season in their 30s even if they’re above-average producers. Therefore, to be considered for a resurgent season, the player has to have had at least two consecutive seasons below 2.0 fWAR/600 in their 30s.
• The resurgent season has to be at or above 3.0 fWAR/600, to give sufficient separation and grounds for improvement, again, for consistency with what Markakis is doing.

In this 2002-2018 period, there have been 2,407 total player-seasons at 30 years of age or older with 200 or more PAs. These player-seasons span only 682 players, so on average, a player that’s lasted into his 30s only really has about 3.5 seasons of being even a half-time-ish player.

Of these 682 players, only 301 had any seasons of 3.0 fWAR/600 or greater. So, right away, we can say that fewer than half of all players that play into their 30s even manage to accumulate an above-average season, on a rate basis, after hitting the three decade mark. After applying the criteria of “two consecutive sub-2.0 fWAR/600” seasons prior to the 3.0 fWAR/600 season,” however, we get... 20. players. Yes, 20. As in, 20 divided by 682 equals three percent. Even among the 301 players that managed to have a 3.0 fWAR/600 season or better, we’re now talking less than seven percent.

So, what Markakis has done this season is pretty rare. Here are his compatriots in achieving this rare feet over the last two decades.

There’s a lot to parse here, but here are the my main takeaways:

• There’s no real position bias here. The number of players at every position is more or less the same.
• Markakis is second only to A.J. Pierzynski in terms of the duration of his decline prior to the resurgent year. Pierzynski is really the exemplar for this exercise; not only did he have a good season after five consecutive barely-average-to-below years, but then he endured two more poor years and had another resurgent season.
• Only two players actually extended their resurgence beyond a single season. Marco Scutaro did something particularly strange — he was pretty much a replacement-level bench player aside from one decent season, and then found another gear in his mid-30s. Even after his two resurgent years, he continued being average-ish until he finally couldn’t manage to stay on the field at age 38 due to a back injury that ended his career. Ramon Hernandez had more of a traditional decline story with a spin — his career year came at age 30, and then he cratered for three seasons before reclaiming a great level of production in a half-time role with the Reds. Then he lost it and was essentially out of baseball two years later.
• Overall, Markakis is fairly in line with his peers here. His decline perhaps started a bit earlier and was a bit less drastic, but his resurgence is very much in line with both the timing and production level seen elsewhere.

So, the answer to the first question is something like: Markakis’ 2018 resurgence is not unprecedented, but it’s fairly rare. The most unique thing about it is probably that he had four below-average seasons before his 2018, while pretty much all other players with similar decline/resurgence patterns (aside from A.J. Pierzynski) generally only tended to have two poor years before a good year, if they were going to have a good year at all.

(By the way, another way to think about this: nearly all players with some below-average years in their 30s don’t ever have another good year. But you probably figured that already.)

What happens after the resurgence?

This is best answered in a table. The results here are fairly instructive for anyone particularly interested in offering Nick Markakis a chance to continue his Braves tenure once his contract expires at the end of this season.

On the one hand, Nick Markakis has made a mockery of his preseason projections. Who’s to say he won’t make a mockery of this table as well? On the other hand, this table is not very favorable. Even among the pittance of players that have managed to do what he’s done, very few of them have gone on to be productive.

If you examine the cells immediately to the right of the first green cell, your results are:

• Below-replacement season: 4
• Below-average season: 5
• Average season: 1
• Another resurgent season: 2
• Could not stay on the field for either health or ineffectiveness reasons: 3

Using the data we have available to us from the above, there’s a small chance (three out of fifteen) that production will be average better. There’s a similar chance said player won’t be able to stay on the field, and a bigger chance that the player will either end up below average, or even below replacement.

If you go two years out, it looks even grimmer:

• Below-replacement season: 3
• Below-average season: 2
• Average season: 1
• Another resurgent season: 1 (Ryan Raburn, with his weird ping-ponging part-time career)
• Could not stay on the field for either health or ineffectiveness reasons: 1
• Failed to make a major league roster and/or retired: 6

Half of these players (seven of 14) failed to even accumulate 200 PAs two years out. Only two of them managed to be average or better.

There’s no doubting that Nick Markakis has been money so far in 2018. But, should being money get him future money? Based on the above, it might be prudent to stay away. In the interim, though, the Braves will take all the resurgent production they can get, especially with Freddie Freeman scuffling a bit.