Our Friend Dave is no Prognosticator.

[brianod]

I try and look past age. Lots of older folks are open to new ideas.

That's actually very funny. I surrender, you win!

[TINSTAAPP]

I still like him more than the writers from Baseball Prospectus. What is the name of the writer who stated that unexpected performances from players like Nolan Reimold was the reason why the Orioles were in first place? I nearly fell out of my chair reading that article. He obviously hadn't watched the Orioles play much.

[Frobby]

This was one of the harder seasons to prognosticate. That said, I think the Yankees have been about what most people expected them to be.

[Tony-OH]

Makes me take take everything "proven" through statistics with a grain of salt. I remember after 20 innings of Bundy, there were statistical arguments that he'll never amount to anything because he's not "missing enough bats". Lol, how's that working out?

Might wanna show us where someone said that. I certainly remember being concerned that his stuff didn't look that good and he wasn't missing bats early on. then the Orioles figured out that he needed at least 4 days off to be effective and his stuff has been better ever since.

For a scouting standpoint, his stuff wasn't that good early on with the short rest. That's why finding the right role for different pitchers can be tricky and why's it's not so easy to just turn a starter into a reliever.

[Tony-OH]

The one thing to remember is the only thing stats tell you 100% is what has happened in the past. Definitely a tool to be used and if used correctly can help contribute to a good scouting report and a good predictor of future performance, but they will never predict the future or tell you what is going to happen.

[DrungoHazewood]

Might wanna show us where someone said that. I certainly remember being concerned that his stuff didn't look that good and he wasn't missing bats early on. then the Orioles figured out that he needed at least 4 days off to be effective and his stuff has been better ever since.

For a scouting standpoint, his stuff wasn't that good early on with the short rest. That's why finding the right role for different pitchers can be tricky and why's it's not so easy to just turn a starter into a reliever.

I said that Bundy was only averaging 6.3 K/9 as a reliever at one point a month or so ago. The implication being that if he transitioned to starter his K rate would likely fall and he might not do so well. The very next post Can_of_corn chimed in that the K rate was climbing, it had been very low in his first 20 innings or so, maybe the K rate wasn't really indicative of talent. A couple posts later I said my post was kind of back-of-the-napkin and might not be entirely valid.

brianod likes to point out my first post in that sequence, as kind of proof that modern analysis is fatally flawed and probably shouldn't be used.

[weams]

The one thing to remember is the only thing stats tell you 100% is what has happened in the past. Definitely a tool to be used and if used correctly can help contribute to a good scouting report and a good predictor of future performance, but they will never predict the future or tell you what is going to happen.

<blockquote class="twitter-tweet" data-lang="en"><p lang="en" dir="ltr">Orioles up to 76% chance to make playoffs on 538. And have best chance to win division at 40%. Hey!! The numbers finally love us!</p>— Josh Sadlock (@JoshSadlock) <a href="

">August 8, 2016</a></blockquote>

<script async src="//platform.twitter.com/widgets.js" charset="utf-8"></script>

And Nate Silver is rarely wrong.

[Can_of_corn]

<blockquote class="twitter-tweet" data-lang="en"><p lang="en" dir="ltr">Orioles up to 76% chance to make playoffs on 538. And have best chance to win division at 40%. Hey!! The numbers finally love us!</p>— Josh Sadlock (@JoshSadlock) <a href="

">August 8, 2016</a></blockquote>

<script async src="//platform.twitter.com/widgets.js" charset="utf-8"></script>

And Nate Silver is rarely wrong.

http://fivethirtyeight.com/features/even-the-orioles-have-a-shot-in-the-al-east/

[LookitsPuck]

Most of us weren't impressed w/ Bundy early on. His command was poor, he wasn't missing bats and hitters were getting on base at a scary clip. And his velocity was down.

That was early on before he was being put more on a starter's schedule. Once that started happening, things started to change. Just look at his game logs:

As a reliever

2 days rest: .385/.448/.500 - .948 OPS, 2 SO in 29 PA, 6 IP, 3 SO/9

3 days rest: .357/.390/.500 - .890 OPS, 9 SO in 60 PA, 13 2/3 IP, 6 SO/9

4 days rest: .154/.214/.231 - .445 OPS, 6 SO in 28 PA, 7 2/3 IP, 7 SO/9

5 days rest: .290/.389/.355 - .744 OPS, 12 SO in 36 PA, 7 1/3 IP, 15 SO/9

Now, those are small sample sizes, I get it. But interesting to see the progression over time. Velocity is better on more rest. Movement is better. As is command.

As a starter:

.177/.223/.396 - .619 OPS, 33 SO in 27 IP, 11 SO/9 with a 2 BB/9.

[countertrey44]

Believe it or not, I believe advanced stats are fine as long as those using them acknowledge that they can be flawed and aren't the end all and be all. So, we are more in the same page then it appears. But, if it helps, you can still pigeon hole me as an ignorant old guy with his head buried in the sand:)

I like to read what others have to say about advance stats, but personally, I only understand the stats that were on the back of baseball cards in the early 80s. Those stats made me fall in love with baseball. I read what others stay the stats mean, but I'm too old and set in my ways to learn another language.

[Phantom]

So wait, a smart person was WRONG about a prediction in something so easy to predict as sports? What a joke that guy must be.

[brianod]

I said that Bundy was only averaging 6.3 K/9 as a reliever at one point a month or so ago. The implication being that if he transitioned to starter his K rate would likely fall and he might not do so well. The very next post Can_of_corn chimed in that the K rate was climbing, it had been very low in his first 20 innings or so, maybe the K rate wasn't really indicative of talent. A couple posts later I said my post was kind of back-of-the-napkin and might not be entirely valid.

brianod likes to point out my first post in that sequence, as kind of proof that modern analysis is fatally flawed and probably shouldn't be used.

This is absolutely not true. I believe advanced stats can play a very useful role in talent evaluation when used correctly and when it is acknowledged that there is no substitution for human judgment. Unfortunately, I see them used incorrectly a lot on this board, such as someone pigeon holing how a young player will develop after limited starts at Bowie or after small sample sizes in the majors. I believe the predictive ability of these stats is much greater for Adam Jones than minor league prospects or rookies. And I do admit it annoys me when I say someone looks like a player and someone rains on my parade with a ton of stats from A ball. It also annoys me when eye tests are scoffed at and these stats are used as absolutes. But, any organization not using them is foolish. They do have a place, just not the only place

[HelenaEngineer]

Might wanna show us where someone said that. I certainly remember being concerned that his stuff didn't look that good and he wasn't missing bats early on. then the Orioles figured out that he needed at least 4 days off to be effective and his stuff has been better ever since.

For a scouting standpoint, his stuff wasn't that good early on with the short rest. That's why finding the right role for different pitchers can be tricky and why's it's not so easy to just turn a starter into a reliever.

I think he may be referring to this article which pretty much says exactly what Brian is talking about http://www.masnsports.com/orioles-buzz/2016/04/ryan-romano-the-dylan-bundy-dilemma.html

[brianod]

I think he may be referring to this article which pretty much says exactly what Brian is talking about http://www.masnsports.com/orioles-buzz/2016/04/ryan-romano-the-dylan-bundy-dilemma.html

Thanks for posting this. A saw can be used to trim a tree or it can be used to chop a tree down. Advanced stats are only useful for someone who knows when and how to use them and that may be more art than science. Obviously, they can be used to prove things that aren't true given the two articles in this thread. But, that doesn't mean I'm against them. That would be like blaming a saw for cutting down a tree