A Close Look at Díaz

[Cowtipper]

So a 40% blown save rate is acceptable as long as the advanced metrics make him look good. Noted.

Organic stats that occur from the natural progression of sport are less flawed than manmade statistics. A manmade stat, more aptly called a 'metric' or 'formula' or 'algorithm', is inherently more unreliable because it is created by man, who is inherently flawed. A double is a double because you see it is a double. To understand WAR, you have to read a long, convoluted explainer. There's a large difference there.

Blowing saves is not doing one's job correctly. Especially when ones job is "closer." Whose job is to "save" games. Mental gymnastics cannot wiggle out of that one.

So there is no trend, by your own admission. Which means there is not a positive trend either. Which should be disconcerting, at $20 million per year. One should expect a positive trend considering his supposed pedigree, and his current pay grade.

Diaz ranks 18th out of 29 in save percentage since joining the Mets in 2019, among pitchers with 50+ saves. That means he's done worse than ~61% of those on the list.



Among pitchers with 100+ saves since 2019, he is SECOND-TO-LAST in save percentage. He has been the second-worst closer since 2019. Since a closer's job is to save games successfully.

[Edgy MD]

So a 40% blown save rate is acceptable as long as the advanced metrics make him look good. Noted.

This is what reductive means.

All statistics are man-made.

[Cowtipper]

Note how I'm actually providing statistics to back my argument and not just snideness. Interesting disparity in our styles.

[Cowtipper]

This is getting boring, I'll let you have the last word.

[Edgy MD]

Cattywampus.

[batmagadanleadoff]

All statistics are man-made.

Not really. A double, for example, is a double. To calculate how many doubles a player hits in a season, you simply have to count the doubles and add them up. The doubles total for any given player is indisputable, and 100% reliably accurate so long as the counting is accurate. There's nothing "man-made"abut a player's doubles totals, even though someone has to physically count them up or tabulate them.

But how accurate or reliable is WAR? It's a calculation, an estimate -- derived from man-made formulas WAR is not a natural stat. The two most influential or widely-used formulations for WAR -- BRef's and Fangraph's, differ. I have no idea which of the two is most preferable, but that they're different is telling. It tells us that WAR is not entirely precise. So there's obviously a built-in margin of error in WAR. By comparison, there's no margin of error in a player's doubles totals, not even a microscopic one, so long as the counting is accurate. Pete Alonso hit precisely 21 doubles last year. That's not an estimate and there's no margin of error in that stat.

[metsmarathon]

Also I called the collapse of deGrom long before anyone else did.

so, like, do you have a history of predicting success for players against which we can judge your clairvoyant abilities?

there are two interesting things about diaz' season so far. the first is that, per fangraphs, he's allowed a lot of soft, pillowy contact.

of all 2024 pitchers with >10 innings pitched, he's got the 5th highest soft contact percentage. that's ordinarily a good thing. 5th lowest average exit velo, too.

the weird thing is that, quite oppositely of the above statement, he's also allowed a very high ratio of home runs per fly ball, of 26.8%. he's 18th worst in that regard. that's, well, usually not a good thing.

his FB% is relatively high, at 43% - which is about what is was in '17, '19, and '21. the HR/FB rates those years was 14.3%, 26.8%, 5.3%, respectively.

nothing particularly jumps out to say that he of course should be bad this year. biggest thing that's noticeable is that hitters are pulling the ball against him a lot more. 48.6% whereas his career average is 37.8%.

otherwise, well, his velocity is down. average FB is at 97.4 mph. in 21-22, it was 90 mph. granted in '18 it was 97.5 so that's not it alone, necessarily. velocity aside, his pitches have very similar movement too. -8.6 to -8.5 inches of horizontal break on the fastball, 7.3 to 7.7 inches vertical. -.1 to +.1 horizontal on the slider, 3.0 to 2.3 vertical.

he's actually got more movement on his pitches this year, they're just a bit slower!

what does it all mean...? dunno. is he declining? is he bouncing back from injury? could it be a mix of both? it could be both.

the soft contact thing gives me hope that the HR/FB rate will right itself. i suspect that as the season goes along, he'll gain more confidence in his leg and his surgically repaired patellar tendon, and might find some more velocity, and look more like the edwin diaz we're used to.

or we're fucked. either is possible.

but! i guarantee you one thing.

edwin diaz will suck, eventually. plan on it.

[metsmarathon]

Pete Alonso hit precisely 21 doubles last year. That's not an estimate and there's no margin of error in that stat.

pete alonso and zach mckinstry both hit 21 doubles last year. who's the better hitter?

[Edgy MD]

But how accurate or reliable is WAR?

I dislike WAR, and I dislike bWAR more than fWAR. I especially dislike how it misrepresents the performance of relievers, whose work is very situation-specific, while all versions of WAR that I know extract all performance from the context in which it occurs and normalizes it with regard to situation.

But whatever bWAR is, it collects a broad range of performance indicators and it's the same thing from year to year, so I took it as meaningfully illustrative of a player's performance from year to year. I don't think it's the useful tool it's presented as in comparing Edwin Díaz 2018 to, say, Dick Groat 1961, but I think it's a very solid bottom line for comparing Edwin Díaz 2018 to Edwin Dîaz 2019.

The same guy in the same position measured by the same stat that compiles from a broad range of indicators (doubles, alone, is pretty singular, and taken alone would indeed suggest that Alonso was as good in 2023 as McKinstry), across consecutive years, controls for a lot of variables.

[Frayed Knot]

OK, I'm going to weigh in on this topic having read the early posts about whether or not Diaz was a much better Mariner than Met and if he is or is not
trending downward. Once the thread turned into having arguments about the arguments my hair started to hurt and I tuned out.


OK, so there is an outlier season for Diaz but it was more 2019, his first in Queens, than it was 2022.
This certainly isn't news to anyone here, we all saw it and it was bad (made worse by the fact that Familia was at least as bad). So while I'm
not trying to deny it happened, I am going to toss it aside for now and see how the comparisons look without it.

So the chart below shows Diaz for his 3 seasons as a Mariner; line 2 is Diaz as a Met minus that 2019 season and the 5.59 ERA; then Edwin's
entire NYM career (3 full seasons plus 2 partial ones) on line three.

	G	IP	ERA	WHIP	ERA+	H/9	BB/9	K/9	HR/9
SEA	188	191	2.64	1.02	156	6.1	3.0	14.2	0.9
NYM - 2019	166	167	2.37	0.99	169	5.7	3.3	15.2	0.6
All NYM	232	225	3.20	1.09	126	6.5	3.3	15.2	0.9

What is clear from the above is that, aside from that one season, his NYM numbers are as good or better than his SEA numbers across the board.
And even with 2019 included, having the worst season in what was essentially the mid-point season of a career doesn't make anything a trend in
either direction. Also, hitting your nadir at age 25 and then bouncing back doesn't indicate anything resembling age related decline.

Now, is he a reduced pitcher on account of the year-long injury? Is the pitcher he's been so far in 2024 the one he's locked into going forward?
While both are possibilities, it's FAR too early to know that and I certainly wouldn't base an answer on the 16 innings he's pitched this year.
The disturbing part so far, which I mentioned during Monday's IGT, is the HRs. But, again, let's wait and see. 4 HRs in 16 innings is close to
the same pace as his 2019 15 in 58 but to call them the same now is to somehow know that this is a trend locked in place and life and sports
don't work that way.