I'll just start now the tourist

Checking out the slums

With my plastic visa

Drinking with my chose

Money, money

Good morning and welcome to episode

3535

Effectively wild the daily podcast from baseball perspectives presented by the play index at baseball reference.com

I'm Sam Miller with Ben Lindbergh from Grantland. Hi Ben. Hello. I use the play index today for a fun thing that I will be writing about for Fox Sports tomorrow

Maybe I'll mention it later. Maybe I won't but I've used it today also for something that I think I'll be writing about for Thursday. What's the topic?

You're not giving yours away. I'll give mine away. I just didn't I mean I didn't want to dive into something that is necessarily but I'm happy it's not a secret.

I'm looking at I was looking at the best relief seasons by various measures. It has all sorts of different ways that you can look at the best relief seasons WPA and RE24

And any stat that you might want to use to evaluate relief seasons. So I was looking to see where Dell and Betoncis ranks and where other people rank.

Interesting. How does Dell and Betoncis rank? I got a read to find out. I see. Did you know I was noticing the other day on play index that

Araldis Chapman has struck out 17 and a half batters per nine and the number two is I think Betoncis who's at 14 and a half by nine.

It's like the gap is three strikeouts for nine. There's no being that gap. Yeah, that's huge. It's incredible. I was looking at MVP voting trends through history and I was just wondering.

I was curious about how how much more closely MVP voting cues to to war nowadays. And I'm not saying it's because war exists. It might be partly because that it might just be because the things that war captures are more valued and more recognized today.

But the correlation between war and MVP votes among players among players who get at least one vote. So this leaves out the hundreds and hundreds of players who don't get any vote.

And also I'm comparing. Well, I also know this is not actually I'll just say this is not a great way to run correlations. For instance, I ran a correlation between war and their actual rankings.

One, two, three, four, five, six and Russell Carlton taught me long ago that you can't compare a stat to ordinal rankings like that. I don't know why, but you can't.

But I did. And the correlation between this between last year and exactly 20 years earlier in 1993 is like double. And that was 93. It's not like 93 was before they had stats or anything like that.

But it was double. It was like a very weak correlation 20 years ago. And now it's a pretty strong one from point three to point six.

...