Harper to Phillies 13yr/330 mil

[SonofaRoache]

57 minutes ago, mqr said:

It's more like 10 times out of a hundred. Or the difference between a .200 hitter and .300 hitter. Last year Harper was on base almost 15 extra times out of 100. Nobody is saying a walk is better than a hit. It's not. It's just better than an out 99% of the time.

Teams that are on base more score more. Therefore they win more.

Your point is valid. The thing is are run producing hits are more valuable than walks. My point is to not just look at OBP but OPS. They had a similar year in that department last year. So one cannot say Harper is a big upgrade over Machado offensively. Harper may have been on base more, but Machado wasnt that far behind offensively overall.

[ptatc]

24 minutes ago, Look at Ray Ray Run said:

Well, OPS is just saying how many bases you can expect to get per at-bat. That means... that it is a very strong correlation since 4 bases = a run, so someone who averages more bases every at bat, contributes more runs.

None of that has anything to do with how strong the predictive correlation is. All that sounds good but until you run the numbers, you don't know.

[SonofaRoache]

24 minutes ago, Eminor3rd said:

Again, outs are never a good thing. 

Sometimes, coaches are willing to accept them as a cost to create a run scoring situation. But in literally every one of those situations, a hit is even better, and in most of those situations, a walk is better. 

It’s a subtle but important distinction from what you’re saying. 

Outs aren't a joy, and coaches prefer hits, bit baseball is a sport where a strong of hits is unlikely, especially with the bottom of the lineup. Coaches sacrificing outs to get runs means they don't like the odds of their batters succeeding without sacrificing. So outs aren't great obviously, but they are the better bet in many cases. 

[ptatc]

14 minutes ago, Balta1701 said:

 

Interestingly, I found a thread that started where someone did the correlations on reddit and wRC+ only had a correlation coefficient with runs scored of 0.83 in one year. It looks like most years it's better than that, but over some long time periods that people tried, wOBA had a higher correlation coefficient than wRC+. I don't see anyone who did just raw OPS in that thread, but raw OPS as I showed above is approximately as good as wOBA and maybe sometimes better. 

So yeah, OPS has a good chance of being more tightly correlated with runs scored than wRC+. 

.83 is pretty strong. It would be good to see if OPS was better.

[Balta1701]

2 minutes ago, Look at Ray Ray Run said:

Ahhh this is where I made my mistake. Its wRC not wRC+. You are correct the adjustment for parks and etc harm it. Similarly to how ops+ is less efficient 

wRC+ is more efficient than OPS+.

if it comes down to it - wRC+ probably does a better job of summing up how truly valuable a hitter is compared to OPS, but the correlation between OPS and runs scored is so strong that you're almost never going to see somewhere where they are a consistent outlier between wRC+, OPS, and runs scored. It's like the difference between walking at 89.9 degrees from north and Due East - you're still going to wind up heading east. So, OPS really does do a great job of summing up a person's performance. 

[Jack Parkman]

Just now, ptatc said:

.83 is pretty strong. It would be good to see if OPS was better.

Click on Balta's link. OPS is .900

[Balta1701]

Just now, ptatc said:

.83 is pretty strong. It would be good to see if OPS was better.

They didn't do it in that one exercise but look above, I've linked other analyses that found it at 0.9 to 0.95.

[Look at Ray Ray Run]

1 minute ago, SonofaRoache said:

Outs aren't a joy, and coaches prefer hits, bit baseball is a sport where a strong of hits is unlikely, especially with the bottom of the lineup. Coaches sacrificing outs to get runs means they don't like the odds of their batters succeeding without sacrificing. So outs aren't great obviously, but they are the better bet in many cases. 

No they're not.

[ptatc]

41 minutes ago, mqr said:

And obviously lineup limitations change the math, but on average a runner on first with 0 outs scores more often than one on second with 1 out who scores more often than a runner on 3rd with 2. Outs, in general, are almost always bad.

Again, it makes common sense but that doesn't meanit's TRUE. Until someone calculates it, you can't definitively conclude this.

[ptatc]

1 minute ago, Balta1701 said:

They didn't do it in that one exercise but look above, I've linked other analyses that found it at 0.9 to 0.95.

Wow. That's excellent when dealing with human subjects.

[ptatc]

4 minutes ago, Balta1701 said:

They didn't do it in that one exercise but look above, I've linked other analyses that found it at 0.9 to 0.95.

I'm surprised batting average is at .79, while HR is .56. 

Have you found anything about runs scored or runs given up and the correlation to wins? What's more important offense or pitching/defense?

[Balta1701]

3 minutes ago, ptatc said:

I'm surprised batting average is at .79, while HR is .56. 

Have you found anything about runs scored or runs given up and the correlation to wins? What's more important offense or pitching/defense?

Here's someone who did the teams for 2002 and found that runs scored has R2 = 0.46 with wins and runs allowed was 0.70. So neither one of them is a great predictor overall but they're both related, with runs allowed being slightly more correlated at that time. 

That gets you almost to the Bill James Pythagorean Wins, of (runs scored)^2 divided by (runs scored^2 + runs allowed ^2) as quite strongly correlated with wins. Basically that's a version of "run differential", which is the thing that translates most strongly to wins. 

[SonofaRoache]

17 minutes ago, Look at Ray Ray Run said:

No they're not.

Then why do coaches sacrifice? Why do hitters move runners over? Why do pitchers bunt when people are on base and no out? 

[ptatc]

5 minutes ago, Balta1701 said:

Here's someone who did the teams for 2002 and found that runs scored has R2 = 0.46 with wins and runs allowed was 0.70. So neither one of them is a great predictor overall but they're both related, with runs allowed being slightly more correlated at that time. 

That gets you almost to the Bill James Pythagorean Wins, of (runs scored)^2 divided by (runs scored^2 + runs allowed ^2) as quite strongly correlated with wins. Basically that's a version of "run differential", which is the thing that translates most strongly to wins. 

Makes sense. I remember when Bill came out with that theory. The .24 differential is significant. In that year defense had a much stronger correlation. It would be interesting to see it year after year. The differential is obviously the better way. However, as a GM, if it came down to adding the hitter or pitcher, the pitcher would be the answer.

In today's  game the answer may be different.

Edited February 1, 2019 by ptatc

[Look at Ray Ray Run]

2 minutes ago, ptatc said:

Makes sense. I remember when Bill came out with that theory. The .24 differential is significant. In that year defense had a much stronger correlation. It would be interesting to see it year after year. The differential is obviously the better way. However, as a GM, if it came down to adding the hitter or pitcher, the pitcher would be the answer.

In today's  game the answer may be different.

Eh, the problem is your giving the pitchers in general the duty of preventing runs while ignoring the fact that pitchers are much more volatile and dangerous commodity...

In today's game, adding the position player is probably the answer if both players are equal in value because of the increased pitched volatility.

[Look at Ray Ray Run]

7 minutes ago, SonofaRoache said:

Then why do coaches sacrifice? Why do hitters move runners over? Why do pitchers bunt when people are on base and no out? 

There are times in baseball in which scoring ONE run is more valuable to the win rate of your team than scoring multiple. In that scenario ONLY should you sacrifice an out. The frequency in which that happens is incredibly rare which is why 99% of sacrifices are bad baseball

[ptatc]

6 minutes ago, Look at Ray Ray Run said:

Eh, the problem is your giving the pitchers in general the duty of preventing runs while ignoring the fact that pitchers are much more volatile and dangerous commodity...

In today's game, adding the position player is probably the answer if both players are equal in value because of the increased pitched volatility.

I don't think the pitchers are more volatile or dangerous. It's more that the analytics are farther along for the hitters than the pitchers. It's more difficult  to quantify pitching because of other variables such as the defense. 

[GenericUserName]

29 minutes ago, Balta1701 said:

if it comes down to it - wRC+ probably does a better job of summing up how truly valuable a hitter is compared to OPS, but the correlation between OPS and runs scored is so strong that you're almost never going to see somewhere where they are a consistent outlier between wRC+, OPS, and runs scored. It's like the difference between walking at 89.9 degrees from north and Due East - you're still going to wind up heading east. So, OPS really does do a great job of summing up a person's performance. 

I think the big difference is that something like OPS tells you what happened while wRC+ can tell you what will happen in different situations. A better hitting environment will result in a higher OPS and more runs, but if they moved to somewhere else they would have a lower OPS even if they stayed the exact same talent level. wRC+ is adjusted so it doesn't match the results of what happened exactly, but it can tell you the difference in quality of player in a neutral environment which is probably better for predicting future performance. Of course expected stats should be even more predictive of future performance which will likely mean they are also much worse at telling you what actually happened.

[Balta1701]

3 minutes ago, ptatc said:

I don't think the pitchers are more volatile or dangerous. It's more that the analytics are farther along for the hitters than the pitchers. It's more difficult  to quantify pitching because of other variables such as the defense. 

Pitchers are 34% more likely to reach the Disabled List and make up nearly 2/3 of the number of days spent on the disabled list (meaning they're more likely to get hurt and when they get hurt they stay on the DL for longer) according to a 2010 study, so it is absolutely true that pitchers are more dangerous due to injury.

[soxfan49]

31 minutes ago, Look at Ray Ray Run said:

There are times in baseball in which scoring ONE run is more valuable to the win rate of your team than scoring multiple. In that scenario ONLY should you sacrifice an out. The frequency in which that happens is incredibly rare which is why 99% of sacrifices are bad baseball

I think it's like 70% where the winning team in the MLB scores more runs in one inning than the losing team scores in the entire game, which backs up your point even more. It's rare that a single run will be the difference in a win unless of course it leads to a big inning, but only having two outs to do so is also rare.

Edited February 1, 2019 by soxfan49

[ptatc]

29 minutes ago, Balta1701 said:

Pitchers are 34% more likely to reach the Disabled List and make up nearly 2/3 of the number of days spent on the disabled list (meaning they're more likely to get hurt and when they get hurt they stay on the DL for longer) according to a 2010 study, so it is absolutely true that pitchers are more dangerous due to injury.

I wasn't taking dangerous as likelyhood of injury. I took it in the context of volatile and dangerous to predict performance. 

I agree that they have a greater likelyhood to get injured. All the more reason to have a greater number of quality ones.

[Jack Parkman]

https://theathletic.com/791063/2019/01/29/sarris-recent-trends-have-flipped-baseballs-conventional-wisdom-on-first-base-and-short/

Look at the aging curve in this article. It suggests that even at age 26, Harper and Machado's best years are already behind them. If that is the case, I really don't know how the players can deal with this in upcoming negotiations. The only answer is to pay players based on potential, and we already saw in the NFL that they don't like that either. I really don't know what the solution to this is for the players. 

[ptatc]

2 minutes ago, Jack Parkman said:

https://theathletic.com/791063/2019/01/29/sarris-recent-trends-have-flipped-baseballs-conventional-wisdom-on-first-base-and-short/

Look at the aging curve in this article. It suggests that even at age 26, Harper and Machado's best years are already behind them. If that is the case, I really don't know how the players can deal with this in upcoming negotiations. The only answer is to pay players based on potential, and we already saw in the NFL that they don't like that either. I really don't know what the solution to this is for the players. 

The solution would be to get to free agency earlier. However, I have no idea what they would need to give up to get that.

[caulfield12]

1 minute ago, ptatc said:

The solution would be to get to free agency earlier. However, I have no idea what they would need to give up to get that.

Then the small market teams losing guys like Lindor after just 5 or 6 years would give up hope.  It would go against the idea of investing time and energy in following the prospects during a rebuild if you were just going to lose them.  There needs to be something like a franchise player tag or max contract or whatever to give those teams a shot.

Perhaps the current freeze is giving those guys like Jiminez and Vladimir Jr. an incentive to sign a huge long term deal before they ever set foot on a MLB field...buying out a year or two of free agency.

[Jack Parkman]

Just now, ptatc said:

The solution would be to get to free agency earlier. However, I have no idea what they would need to give up to get that.

With most players making their MLB debut at age 22-25, I don't really think there is much they can do. Teams aren't giving up control immediately after a player's rookie season. That is the only way to get them paid during their most productive years. That aging curve is so crazy that there is no practical way to get around it. Teams aren't going to want to pay for potential or decline. If they argue that, there is pretty much zero counter argument from the players standpoint. By the time they've established themselves enough, they're already depreciating assets. It is a losing argument. I think the players have to negotiate a slice of the pie, along with giving up a cap and floor. Otherwise, they have zero leverage.