So much of what we do is try to tell the future. So much of what we do is get the future wrong! But still we keep on plugging away, with any noteworthy trade, or any noteworthy signing. One noteworthy signing last week reunited the Yankees and Aroldis Chapman, with the closer getting a five-year guarantee. The history of relievers getting long-term deals isn’t great, but one thing we can’t ignore is that Chapman is kind of extraordinary. By some measures he’s even unique. Baseball doesn’t have an extended track record of Aroldis Chapmans (Chapmen?). Projecting his future is tricky, as it always is.

We know that, if Chapman stays the same, he’ll remain nearly unbelievable. That much, he’s proven. We know that he could snap something after any given pitch. That much, countless other pitchers have proven. But what else might we able to say?

One issue that’s come up is imagining Chapman throwing at lower velocities. Every pitcher eventually loses his zip. Some start at 25, and some start at 35, but it’s a guarantee. There’s no getting around it. For Chapman in particular, there’s one positive sign. Last year, his average fastball was 100.4 miles per hour. That’s the highest mark of his career, and he was around 98 in his first full season. So Chapman hasn’t lost anything yet. Maybe he’ll continue to hang around triple digits for a while to come.

Inevitably, though, there will be decline. And what then? To attempt to answer that, I want to try something. How might Chapman perform at lower velocities? Well, we can look at how he’s already performed at lower velocities.

Tables, ahoy! I looked at Chapman’s game log from 2012 to 2016. For every single game, I grabbed Chapman’s average fastball velocity, and then it was a matter of putting numbers together. In this first table, Chapman’s numbers are split into six groups.

That’s fairly informative already, but let’s increase the sample sizes. We’ll do that by condensing from six groups to three.

Surprise! When Chapman has thrown his hardest, he’s been his best. He hasn’t allowed many hits, and he’s allowed only two home runs. The home-run category here might be the most dramatic, but it’s also the noisiest, since Chapman hasn’t allowed many homers overall. Even when Chapman has worked around 96-97, he’s still been a strikeout machine. For reference, he’s a top-10 table since 2012, looking at strikeout rate minus walk rate. Chapman leads all pitchers, overall, but I took him out and replaced him with the three versions of himself.

The hardest-throwing Chapman is in first, easily. The medium-throwing Chapman is in fourth. The slowest-throwing Chapman is in ninth, out of 776 pitchers. Ken Giles just missed this table. So did Sean Doolittle. And maybe that’s older Chapman: some kind of healthier Sean Doolittle equivalent. Doolittle has sometimes given up his home runs, and he hasn’t been considered one of the highest-tier relievers, but he has been excellent when he’s pitched. So Yankees fans can be somewhat encouraged.

Aroldis Chapman, at some point, will lose some velocity. Aroldis Chapman, at some point, will get worse. But, of course, he has a lot of velocity to give. And when he’s worked at lower speeds before, he’s been only somewhat less incredible. Chapman comes with a lot of what you might consider wiggle room.