Anyway, in hopes of helping us see how the overall ratings field has changed, I took a standard set of cuts on ratings across the league by going to the game, selecting “Statistics” and then “sortable stats.” At this point I used a custom view that includes all the core ratings, then exported them into a report (that I then rolled into Excel to do some basic slicing and dicing.
To give you a basis of where the conversion starts from, you could recall a bit I posted on ratings right after our conversion to relative ratings, a season and a half ago.Remember, these are relative ratings. The root core ratings are what goes into the game’s results engine, so those will be different (and really should be unchanged by the conversion itself).
Note, also, that these are ratings of players who created stats in 2038, and where their ratings are at this time (late November game time).
Here are the overall ratings and potentials:
Discussion:
Overall, I think I’m pretty happy here. In the past, I think we’ve kind of said 55 has been considered “league average.” Now it looks like it’s 45, which really feels okay. Your mileage may vary, of course. I mean, I wish it would stabilize at 50, but I can also take a step back and realize these are “OOTP overall ratings” which, as Ted showed us, do not do a great job of presenting platoon values, so if I mentally shift this up a notch for platoon advantages, I’m about in the right spot.
I note also that there exists a chance that this curve shifts a little earlier in the year vs. later in the year, depending on the players who’ve been called up or hurt or whatnot. So, yeah, to restate, I’d like it to be less spikey at 45 than it is, but as long as we understand the facts behind the chart there are things to like here.
Another thing I like about this plot is that the talent scale on the right side is less choppy and seems to have fewer guys in the 75 and 80 range than we did after the v19 move to relative ratings.
Still not perfect, but it seems maybe more perfect than before. I’m certainly open to disagreements there, though. In some ways looking at these charts are like going to the art museum and defending why one painting is better than another.
Anyway, let’s move on to the core ratings—which to me are much more valuable anyway.
A Look at Components:
Discussion:
With five elements, this is a bit more of a messier chart. That said, taking a moment to follow the individual skills gives you a feel for where the averages and scatter are. For example, only in the POWER rating do we see substantial “1” and “2” ratings, though the skill seems to peak clearly at “6” and all off with a right skew toward “7.” GAP kind of follows POWER, but is perhaps a bit smoother.
CONTACT, however, while also peaking at “6” falls off more heavily to the left. In fact, “4” “5” and “6” are the most common ratings. EYE is flat at “5” and “6” and skews left. AVK peaks at “6” and skews right.
So there are differences, yes. It’s still probably fair as a rot-gut rule of thumb to say that a league-average component rating is “6,” but operationally, you’d probably be wrong with EYE.
Anyway, I’m not sure how much you want to bake this thinking into your ideas, but there it is.
Again, I think that overall these are a bit nicer, though actually pretty close to the 2037 data. Contact, in particular, seems a little more…appropriate?
I dunno. Again, different strokes for different folks, maybe.
As I know OOTP did some real work on pitcher valuation, I’ll be interested to see how the pitcher ratings come out in the BBA world.