A Realistic View at 2019: Part I

[Kimmi]

I think stats are very useful, as long as you understand how to use them. Stats are necessarily based on past performance. This means that stats have limited predictive ability. For example, people often say a player will regress to the mean. That may be true of certain types of statistics, but it is incorrectly applied to player performance. An injury affects performance which affects statistics. This mean the statistics should reflect the physical condition of the player. It is the physical performance driving the statistics, and that always needs to be taken into account when looking at the numbers. Players don't regress just because regression is a statistical concept.

 

Some stats, like OPS are just bad stats. You shouldn't add together averages that have nothing to do with each other and that are calculated differently. And one of those averages, SLG, is a very flawed stat by itself. It assigns a value to each hit based only on the number of bases touched on the hit. Is that really the true measure of power hitting? Batting average is scoffed at, but that isn't because it assigns value to hits. It has the simple job of calculating the average of how many at bats produce a hit. And if that is all it is used for, it does a pretty good job of that one aspect of the game.

 

It is really all the statistical shortcuts that are the weakness of stats. Sabermetrics came along to improve that. But on this forum I would say that sabermetrics aren't really used, and I include myself.

 

On your first paragraph, I agree completely.

 

As far as OPS and SLG are concerned, I agree that they are flawed, as are all stats, but I don't think that makes them bad stats. IMO, OPS is a very good stat with regard to its correlation to runs scored and its simplicity. BA is quite fine, though I think OPS and many other stats are better in assessing a batter. As you mention, understanding what the stat is trying to measure and understanding its limitations are key.

 

I use sabermetrics all the time, it's just that no one ever listens to me. I also often use the more familiar stats due to their simplicity and familiarity.

[moonslav59]

Most of us still believe the Sox have a solid chance for a wild card slot this year so want us to continue to put our strongest team forward and not start thinking of rebuillding.

 

The move of Leon back and Hernandez down is a no brainer.

Price should be back for Toronto and at that time sending Josh Smith down makes sense.

I'm guessing Eovaldi comes back some time in early June and at that time possibly we lose Brewer.

 

I'm wondering if briging Pedey back at all makes any sense. He appears to be fragile and is not likely to shake the rust off quickly. We have Chavis who while not a good fielding 2nd baseman is at least an acceptable hitter if not more and he is young and durable. Holt is another one of those fragile sorts that would take time to get up to speed and may not add more value than a Lin would.

 

Bringing Johnson back is as a 6th pitcher to be used in spots. He should be given plenty of time to get his act together. He certainly is no answer for our pitching limitations.

 

What I listed were not my opinions. They were what I expect might happen based on what I know at this point in the season.

 

Pedey will be called up, if he gets healthy enough. I'd probably rather have Nunez or Pearce on the 25 man roster than Pedey, especially if it means we have to DFA one of them to make room for the "fragile" Pedey. I'd rather keep Holt over Pedey, too, but we are not going to DFA Pedey or keep him down against his will.

[moonslav59]

The way that Nomar made that play from the SS hole was spectacular. He would somehow twist and throw to first base while he was his momentum continued toward 3B. It was amazing how much he got on those theory and the accuracy.

 

On defense, all flash and little substance.

 

On offense, he was a god.

[Kimmi]

Batting average tells accurately what percentage you hit the ball safely in all your at bats which you hit the ball either for a hit or an out. That is what at bats are considered in baseball and everyone is aware of that fact. Hitting is so difficult that slight variations in average indicate wide variation in the ability to hit safely.

 

SLG is flawed and OPS more flawed. And it appears that not everyone is aware of that.

 

OPS is flawed. So is BA. OPS has a stronger correlation to run scoring than BA does. Also, OPS has a stronger year to year correlation. IMO, that makes it a better stat.

[moonslav59]

On your first paragraph, I agree completely.

 

As far as OPS and SLG are concerned, I agree that they are flawed, as are all stats, but I don't think that makes them bad stats. IMO, OPS is a very good stat with regard to its correlation to runs scored and its simplicity. BA is quite fine, though I think OPS and many other stats are better in assessing a batter. As you mention, understanding what the stat is trying to measure and understanding its limitations are key.

 

I use sabermetrics all the time, it's just that no one ever listens to me. I also often use the more familiar stats due to their simplicity and familiarity.

 

If they created a new metric whereby they counted OBP as 60% of "OPS" and SLG% as 40% of "OPS", it would be more accurate, but hard for many to understand, so we don't do it.

 

3 x OBP + 2 x SLG = Adjusted OPS

5

 

[Kimmi]

OPS is a good measure of offensive productivity. However , it is not a true stat . It is a concoction. Someone had the idea to add two stats together and create a new one . Now , it is pretty much accepted. I don't like the fact that it ignores stolen bases .

 

There are stats that include stolen bases and base running in general, and give you the overall offensive production of and player.

[Kimmi]

...and SLG% > BA

 

OPS> OBP> SLG> BA

 

And wOBA and wRC are even better than OPS.

[Kimmi]

WAR is not that complicated if you accept that its intent is to measure how many RUNS a player produces or prevents above what a replacement player would.

 

WAR is derived from RAR - runs above replacement.

 

Obviously there are a lot of moving parts involved in trying to calculate runs produced and prevented.

 

I'm not interested in all the minutiae of how the calculations are done. I'm satisfied that the metrics folks are doing the best possible job they can. I also accept that there are going to be flaws and anomalies.

 

^^ Full fledged stat geek.

[Bellhorn04]

And wOBA and wRC are even better than OPS.

 

It's when the statistical acronyms start having both upper-case and lower-case letters that a lot of people get scared off.

[Kimmi]

It's the "lot of moving parts" that bothers me. Whenever I have tried to follow a detailed explanation, I always run into areas that involve 'guesswork,' 'the eye test' (or variants), unstated assumptions, various forms of arbitrariness. (e.g., as in BA, which of course, I accept for what it is; what could be more idiotic statistically than failing to credit a bunter for a sacrifice if you think he was trying to get a hit; it would be like refusing to grant an S to the guy who hits a fly ball because you determined that he was actually trying to hit a home run). I'm not questioning the good faith of the statisticians (although I probably should, since some of them are doubtless working for agents); I'm just questioning their abilities to produce a system that doesn't incorporate the same flaws much simpler systems have.

 

The goal of the statisticians is to eliminate as many flaws as possible and to get answers. They are not out to try to prove anyone wrong, though that has happened quite a bit, nor are they out to dupe anyone.

[Kimmi]

If they created a new metric whereby they counted OBP as 60% of "OPS" and SLG% as 40% of "OPS", it would be more accurate, but hard for many to understand, so we don't do it.

 

3 x OBP + 2 x SLG = Adjusted OPS

5

 

 

There actually is a stat, GPA, that has OBP and SLG broken down correctly by percentages.

 

Sadly, it has not caught on.

[Kimmi]

It's when the statistical acronyms start having both upper-case and lower-case letters that a lot of people get scared off.

 

LOL That is a scary thought.

 

Actually, acronyms in general are a scary thought.

[moonslav59]

And wOBA and wRC are even better than OPS.

 

Agreed, but they have not caught on, so not many would understand, if we used them.

[dgalehouse]

Here is the most important stat : Last year on this date , the Sox were in first place with a record of 31 - 15. Today we are in third place with a record of 23 - 22 . Furthermore, we are 3 -6 against the Yankees , Astros and Rays , arguably the only good teams we have played . What is the difference this year and what can be done to change things . ( The good news is that there are so many crappy teams in the league that reaching the post season is still a good bet . )

[harmony]

LOL That is a scary thought.

 

Actually, acronyms in general are a scary thought.

Kimmi, in the discussion of old school versus analytics, I highly recommend The Only Rule Is It Has To Work: Our Wild Experiment Building a New Kind of Baseball Team by Ben Lindbergh and Sam Miller, the former editor in chief and current editor in chief at Baseball Prospectus.

 

The stat geeks were given nearly full rein to make personnel and strategy decisions for the Sonoma team one year in an independent California league. The read is quite entertaining on top of being informative.

Edited May 19, 2019 by harmony

[oldtimer]

Here is the most important stat : Last year on this date , the Sox were in first place with a record of 31 - 15. Today we are in third place with a record of 23 - 22 . Furthermore, we are 3 -6 against the Yankees , Astros and Rays , arguably the only good teams we have played . What is the difference this year and what can be done to change things . ( The good news is that there are so many crappy teams in the league that reaching the post season is still a good bet . )

 

Here is a surprising stat. Of the regullar players with a significant number of plate appearances (I rule Nunez out of that list), who has struck out the most and who the least.

 

The surprise is our leadoff hitter Beni has gone down 45 times. Equally surprising is that our catcher, Vazquez has only struck out 20 times. I wouldn't have guessed that at the beginning of the season.

[moonslav59]

Here is a surprising stat. Of the regullar players with a significant number of plate appearances (I rule Nunez out of that list), who has struck out the most and who the least.

 

The surprise is our leadoff hitter Beni has gone down 45 times. Equally surprising is that our catcher, Vazquez has only struck out 20 times. I wouldn't have guessed that at the beginning of the season.

 

Beni has way more PAs than most players on the Sox. Here are the K% for Sox players (65+ PAs)

 

%

36 Pearce

30 JBJ

27 Chavis

24 Beni

21 Moreland

18 Bogey

17 Nunez

17 Vaz

15 Devers

15 Betts

15 JD

 

[jad]

Of course. And I'm not saying that everyone has to embrace the new stats. I'm really not. I firmly believe, though, that if one is going to debate on a baseball forum, being able to support one's opinion with numbers goes a lot further than anecdotal evidence.

 

I do understand that not everything can be supported by data. I have made arguments that I cannot support by data. However, if someone showed me strong statistical data that went against my opinion, I would not brush the data off as being wrong or useless. I would rethink my opinion and consider that I might be wrong.

 

It's okay if you don't embrace analytics. It's not okay, IMO, to say that you embrace analytics or see the value in them, use them when they support your opinion, but then blow them off as useless when they say something that contradicts your beliefs or opinions. And I don't mean 'you' personally.

 

Did you read the depressing article in The New Yorker, reviewing studies on the relation of evidence to one's opinions? The studies all set up experiments where the subjects were asked to form an opinion (or choose between alternatives) of some artificial problem. They were then told OOPS, that was a mistake: here's what the situation really is. Or perhaps they were presented then with hard evidence that went against their initial opinions. What the researchers found consistently was that when subjects were presented with evidence that absolutely contradicted their initial choice or opinions, far from modifying those views, they became even MORE ADAMANT about them.

[notin]

Did you read the depressing article in The New Yorker, reviewing studies on the relation of evidence to one's opinions? The studies all set up experiments where the subjects were asked to form an opinion (or choose between alternatives) of some artificial problem. They were then told OOPS, that was a mistake: here's what the situation really is. Or perhaps they were presented then with hard evidence that went against their initial opinions. What the researchers found consistently was that when subjects were presented with evidence that absolutely contradicted their initial choice or opinions, far from modifying those views, they became even MORE ADAMANT about them.

 

 

That is called The Backfire Effect. The more you disprove someone’s belief, the stronger their belief becomes...

[S5Dewey]

Wrong thread. This is what happens when one tries to post and watch the game at the same time. LOL Edited May 19, 2019 by S5Dewey

[jad]

That is called The Backfire Effect. The more you disprove someone’s belief, the stronger their belief becomes...

 

Thanks. I thought I knew most logical fallacies and related socio-psychological principles. Alas, it explains far too much of the political situation today than I would wish!

[S5Dewey]

Thanks. I thought I knew most logical fallacies and related socio-psychological principles. Alas, it explains far too much of the political situation today than I would wish!

 

That was my thought too, but I didn't want to get into it.

[moonslav59]

Nice win yesterday. Sale was not at his best, but we squeaked enough runs across to get the W.

[mvp 78]

@OverTheMonster

 

Red Sox option Josh Smith and Óscar Hernández to make room for David Price and Sandy León

[moonslav59]

@OverTheMonster

 

Red Sox option Josh Smith and Óscar Hernández to make room for David Price and Sandy León

 

Those were the easy choices. It's gets a bit harder when the next guys are ready....

 

Pedroia (Go to 12 pitchers: Weber or Brewer to AAA or DFA/trade Thornburg)

Holt (Must decide who to keep- Nunez, Holt or Pearce)

Johnson (Weber or Brewer to AAA or DFA/trade Thornburg)

Eovaldi (Weber or Brewer to AAA or DFA/trade Thornburg)

 

Final 100% healthy 25 man roster:

 

SP (6): Sale, Price, Porcello, Eovaldi, ERod, Johnson

RP (6): Barnes, Workman, Brazier, Walden, Hembree, Velazquez

 

C (2): Vaz & Leon

1B (1): Moreland

2B (1): Chavis

3B (1): Devers

SS (1): Bogaerts

LF (1): Benintendi

CF (1): JBJ

RF (1): Betts

DH/OF (1): JD

Bench (3): Holt and 2 from: Nunez, Pedey or Pearce.

 

A look at the final 3 slots:

 

Holt: He can play 1B, as can Chavis, and he can play a decent OF, so I slot him ahead of Pearce & Nunez.

 

Nunez: He needs to start hitting, or he may be the odd man out, but he's more reliable than Pedey (which isn't saying much). He can also play a better OF than Pearce, when healthy.

 

Pedey: He will not be DFA'd. He may retire, if it comes down to him not being able to play more than 2 times a week. I think he stays over Pearce, if he can show he can play 3+ times a week, if needed.

 

Pearce: If he doesn't start hitting, I think he'll be the odd man out. Even if he does hit, we may keep Holt, Nunez & Pedey over him.

 

Tough call.

 

(DFA/Traded: Thornburg)

 

Edited May 20, 2019 by moonslav59

[mvp 78]

Pedroia - go to 12 pitchers and drop the lowest man on the totem pole (Weber)

Holt - DFA Holt

Johnson - drop Brewer

Eovaldi - drop Thornburg (good chance some other pitcher is injured before then though)

[moonslav59]

Koji Retires in Japanese League.

[moonslav59]

Pedroia - go to 12 pitchers and drop the lowest man on the totem pole (Weber)

Holt - DFA Holt

Johnson - drop Brewer

Eovaldi - drop Thornburg (good chance some other pitcher is injured before then though)

 

I have so little faith in Brewer and Thornburg that I'd probably keep Weber on the big club for a few more shots at showing he's better.

 

I'd keep Holt over Nunez and Pearce.

[mvp 78]

I have so little faith in Brewer and Thornburg that I'd probably keep Weber on the big club for a few more shots at showing he's better.

 

I'd keep Holt over Nunez and Pearce.

 

Holt hasn't been all that productive since 2015. He's constantly injured and doesn't have the upside of Nunez or Pearce. It's an easy call for me. I also think you could sneak him through waivers and get him to AAA.

[notin]

Holt hasn't been all that productive since 2015. He's constantly injured and doesn't have the upside of Nunez or Pearce. It's an easy call for me. I also think you could sneak him through waivers and get him to AAA.

 

 

Do they need to sneak him through waivers? According to soxprospects.com, he still has one option left..