Photo

Bundy Traded to Angels for 4 Minor League Pitchers


  • Please log in to reply
203 replies to this topic

#161 weird-O

weird-O

    All Star

  • Members
  • PipPipPip
  • 1,206 posts
  • LocationI'm here from downtown, I'm here from Mitch and Murray.

Posted 10 December 2019 - 04:46 PM

Could one of you guys with the positive outlook explain to me what the plan is?

My impression, is that the new FO is focused on creating an actual analytics department, and building a network within LatAm that will allow them to join the other 29 teams in securing commitments, from quality players, to sign with the O's, when they're old enough. Those things are a sharp departure from how Peter ran the team. And it sets the O's in a direction that will allow them to be on equal ground, with the rest of the sport. They need to execute, but from what we're being told, they're the guys that can get the job done.    

 

As for the battle between the "they can compete now and build the future" and "This plan is the only way" camps, I don't have anything new to add to that debate. I see good points brought up by both sides. 


  • BSLMikeRandall likes this

Good news! I saw a dog today.


#162 BSLChrisStoner

BSLChrisStoner

    Owner

  • Administrators
  • 116,901 posts

Posted 10 December 2019 - 04:50 PM

I'm 100% fine with this approach.  There is value, both to the team and entertainment wise, in watching young players with promise even if they are collectively awful.  If a team full of guys like Hays and Akin and Harvey goes out and loses 107 games that is much more palatable to watch than a team full of guys like Rio Ruiz and Mike Wright.


I'd be pretty excited for '20, if we we're operating like '89 and putting our best internal talent on the field. 

Do understand '20 economics and team control, etc etc. 



#163 Mackus

Mackus

    HOF

  • Members
  • PipPipPipPipPip
  • 42,640 posts

Posted 10 December 2019 - 05:01 PM

You're overstating the difference between 1-2 a bit.  2 still has a sizable advantage over 6-10.  It doesn't flatten out until like pick 3-4.   https://fivethirtyei...its-usefulness/ . 1's expected value is 12ish WAR, 2 is at 10ish, and 6-10 are worth like 3-6.  

 

I've been looking at career WAR from BB-ref, typically using the drafts from 1980-2004 but I can adjust that to look at any range (I haven't updated the data since mid-2018, so anyone active will be shortchanged, which is why I cut things off back in '04 rather than a bit later, to avoid guys still playing).  Data in your link above is for 1965-2007 and is for first 6 years of service.  The trends are similar, regardless of the numbers and data sets.

 

Look at the scale of the article you're talking about.  12 WAR in the first 6 years versus 8 or 6 or 4.  Who cares?  That's basically the difference between an average starting caliber player and a borderline starting caliber player.  The average value of any top pick is not much, because there are so many misses at every spot in the draft.  

 

In addition to looking at average value, we should also look at median value and what percentage of players reach a certain threshold.  The "lottery ticket" aspect.  Again, #1 has a noticeable advantage over #2 (while still not what anyone would consider "good" odds), but by the time you're at #2 it really levels off whether you're looking at 20 career WAR, 30, 50...compared to picking later in the top-10.  It's possible teams have discovered something in scouting in the last 15 years and the more recent results will completely confound what happened historically before, but I doubt it.  I can't counter that sort of argument because it's an unproveable statement.



#164 McNulty

McNulty

    la cerveza está muy fría

  • Members
  • PipPipPipPipPip
  • 7,790 posts
  • LocationBS

Posted 10 December 2019 - 05:07 PM

I've said this before, but the bonus pool itself is probably more important than draft position.  

 

The Orioles crying poor is *utter* nonsense, in any context.  They are *lying* to you when they say we are small market, its not a thing these days.  These are billionaires, who have stadiums subsidized and are flush with cash from the 1.5 billion dollar streaming boom.  

 

I am very much in favor of sucking horribly again this year, but they should be using the low payroll to leverage taking on salaries for prospects.  Giving away Villar was unforgivable, truly.  You pay 80% of his salary and you get a real piece back from someone; OR you can flip him at the deadline.  Watch the Marlins do exactly that.  


  • BSLChrisStoner, Ricker Says, Mackus and 1 other like this

@fuzydunlop


#165 Mackus

Mackus

    HOF

  • Members
  • PipPipPipPipPip
  • 42,640 posts

Posted 10 December 2019 - 05:10 PM

I've said this before, but the bonus pool itself is probably more important than draft position.  

 

Yes, this is a fair point.  It's not what anybody is arguing for when they say they want the #1 pick, but getting a bigger bonus allocation does at least give you the option of spreading your money out differently which could be more beneficial and would need to be accounted for differently than simply comparing the career performance of players by draft position.

 

If you spend slot on your top pick, then that potential option is gone and the question reverts to it's more simple form.



#166 BSLMikeRandall

BSLMikeRandall

    Sr. Ravens Analyst

  • Members
  • PipPipPipPipPip
  • 20,117 posts

Posted 10 December 2019 - 05:25 PM


Seriously, The real plan is to build through the draft and player development. I have a lot of faith in some of the people that we've brought in.


I have faith too. I love the forward thinking they bring. But you can do these things without trying to lose 130 games. You can keep a 30/30 guy around and still pick high in the draft. The players at the ML level have nothing to do with development at the lower tiers.
@BSLMikeRandall

#167 Nigel Tufnel

Nigel Tufnel

    HOF

  • Members
  • PipPipPipPipPip
  • 4,519 posts

Posted 10 December 2019 - 05:47 PM

Could one of you guys with the positive outlook explain to me what the plan is?

 

Interesting to see that the O's cancelled Fanfest as soon as Ynoa signed to plan in Japan.  Coincidence?



#168 BSLMikeRandall

BSLMikeRandall

    Sr. Ravens Analyst

  • Members
  • PipPipPipPipPip
  • 20,117 posts

Posted 10 December 2019 - 06:13 PM

Furthermore, if we’re worried about blocking guys, there is an argument that Davis is blocking guys. The money sucks, dropping him is like setting it on fire. But if we’re thinking about the long term, someone else is better served to develop in his spot.
@BSLMikeRandall

#169 BSLSethBondroff

BSLSethBondroff

    BSL Analyst

  • Members
  • PipPipPipPipPip
  • 13,952 posts
  • LocationEldersburg, Md

Posted 10 December 2019 - 06:19 PM

Furthermore, if we’re worried about blocking guys, there is an argument that Davis is blocking guys. The money sucks, dropping him is like setting it on fire. But if we’re thinking about the long term, someone else is better served to develop in his spot.

This i absolutely agree with. Mancini might even be more likely to stay if we weren't stuck with Davis. 


@beginthebegin71

#170 BSLRobShields

BSLRobShields

    BSL Analyst

  • Members
  • PipPipPipPipPip
  • 87,188 posts
  • LocationBaltimore

Posted 10 December 2019 - 06:33 PM

This i absolutely agree with. Mancini might even be more likely to stay if we weren't stuck with Davis.


All the more reason to keep Davis then.
@BSLRobShields

#171 dude

dude

    HOF

  • Members
  • PipPipPipPipPip
  • 6,916 posts
  • LocationColumbus, GA

Posted 10 December 2019 - 07:28 PM

If you're not using WAR, how are you estimating the effect on W/L that adding league-average journeymen will have?  It's at least a rough approximation, no?

 

No.  WAR is not a predictive tool for a teams win total, especially the way you are trying to use it.



#172 dude

dude

    HOF

  • Members
  • PipPipPipPipPip
  • 6,916 posts
  • LocationColumbus, GA

Posted 10 December 2019 - 07:29 PM

Seriously, The real plan is to build through the draft and player development. I have a lot of faith in some of the people that we've brought in. 

 

Seth, at what win total are the Orioles no longer allowed to draft or execute player development? 

 

We should win as many games as we can and stay just under that number.



#173 BSLSethBondroff

BSLSethBondroff

    BSL Analyst

  • Members
  • PipPipPipPipPip
  • 13,952 posts
  • LocationEldersburg, Md

Posted 10 December 2019 - 07:29 PM

All the more reason to keep Davis then.

For now, yes, I meant in as if the Davis deal never happened terms. 


@beginthebegin71

#174 hallas

hallas
  • Members
  • 497 posts
  • LocationSomewhere around LA

Posted 10 December 2019 - 07:53 PM

No.  WAR is not a predictive tool for a teams win total, especially the way you are trying to use it.

 

What is it then?  WAR has around a .86 correlation coefficient with wins, so if you're attempting to predict future WAR (and plenty of prediction models attempt to predict WAR) you can also use it to ballpark team wins.


  • BSLChrisStoner likes this

#175 dude

dude

    HOF

  • Members
  • PipPipPipPipPip
  • 6,916 posts
  • LocationColumbus, GA

Posted 10 December 2019 - 09:46 PM

What is it then?  WAR has around a .86 correlation coefficient with wins, so if you're attempting to predict future WAR (and plenty of prediction models attempt to predict WAR) you can also use it to ballpark team wins.

 

That's looking backwards.  That result wasn't the only result for that season.

 

The Orioles won 47 games in 2018....they could have taken the exact same team and won 45 more games, because the individual and collective results could be dramatically different.  The same in 2014.  That wasn't a 96 win team, you could take the exact same team and win 70 games in another season.

 

WAR is an inconsistent stat (measurement, depending on who is doing it) that uses a generic application (not team based in the coefficients) and assumes any number of things about the consequence of baserunning and defense that may or may not be true, utterly ignores all of the real influences that impact winning including collective relationships, team dynamics, coaching and other effects (that can impact individual performance), while also ignoring the Division and League environments that you have to play in.

 

When you look backwards, you are only describing what happened.

 

What happened (one instantiation) may have little relationship to what will happen.



#176 hallas

hallas
  • Members
  • 497 posts
  • LocationSomewhere around LA

Posted 11 December 2019 - 12:16 AM

That's looking backwards. That result wasn't the only result for that season.

The Orioles won 47 games in 2018....they could have taken the exact same team and won 45 more games, because the individual and collective results could be dramatically different. The same in 2014. That wasn't a 96 win team, you could take the exact same team and win 70 games in another season.

WAR is an inconsistent stat (measurement, depending on who is doing it) that uses a generic application (not team based in the coefficients) and assumes any number of things about the consequence of baserunning and defense that may or may not be true, utterly ignores all of the real influences that impact winning including collective relationships, team dynamics, coaching and other effects (that can impact individual performance), while also ignoring the Division and League environments that you have to play in.

When you look backwards, you are only describing what happened.

What happened (one instantiation) may have little relationship to what will happen.

1: How is Steamer or PECOTA looking backward when they say “x player is predicted to accrue 5 WAR in 2020 as the median prediction.”
2: the r value for WAR to team wins is already quite high, so while it’s possible that you’ll have a team that doesn’t add up to the sum of its parts, that result is more likely to be an outlier. When predicting the future, you generally take the non-outlier outcomes.

#177 dude

dude

    HOF

  • Members
  • PipPipPipPipPip
  • 6,916 posts
  • LocationColumbus, GA

Posted 11 December 2019 - 12:29 AM

1: How is Steamer or PECOTA looking backward when they say “x player is predicted to accrue 5 WAR in 2020 as the median prediction.”

 

I'm happy to discuss this as much as you want.

 

I don't understand why you don't understand your 1. above.  It's calculation off the prediction numbers.  a) there's nothing that says those numbers are right (they're just another layer of projections) and b) adding WAR projections to the win totals of the previous season still doesn't mean anything, especially when you move to the extremes like the Orioles 2018 season.

 

WAR can still be used for stuff....but that's not it.



#178 JeremyStrain

JeremyStrain

    Moderator

  • Moderators
  • 11,309 posts
  • LocationLanham, MD

Posted 11 December 2019 - 12:43 PM

1: How is Steamer or PECOTA looking backward when they say “x player is predicted to accrue 5 WAR in 2020 as the median prediction.”
2: the r value for WAR to team wins is already quite high, so while it’s possible that you’ll have a team that doesn’t add up to the sum of its parts, that result is more likely to be an outlier. When predicting the future, you generally take the non-outlier outcomes.

 

I hate getting in these, especially when I don't have a dog in the fight, I hate when these kinds of half predictions come out...I know they can SAY oh based on previous production this is what we expect, but the real question, and a better statistic to use hand in hand with this would be HOW often are those statistics correct, even within one standard deviation?

 

Now if you said, we predict him to be worth 5 WAR in 2020, and based on previous comparable predictions, or confidence on that is 80%, I say that's about as good as an educated guess as you are going to get. If you said we predict 5 WAR and our confidence is 15% it's a much different story right?

 

I'm just used to seeing tables for these kinds of predictions, it'll be like:

>5 WAR - 40%

>4 WAR - 20%

>3 WAR - 10%

<2 WAR - 30%

 

That gives you an overall better feel of how likely those predictions are to hit...or at least help temper expectations.


@JeremyMStrain

#179 McNulty

McNulty

    la cerveza está muy fría

  • Members
  • PipPipPipPipPip
  • 7,790 posts
  • LocationBS

Posted 11 December 2019 - 01:41 PM

Not a Dan Connolly fan but he wrote an article today heavily implying that the Orioles refuse to leverage teams that want to dump salaries for prospects, like the Cozart trade.

 

*Really* disappointing to hear this.  https://theathletic....al-constraints/


  • Ricker Says and Mackus like this

@fuzydunlop


#180 Mackus

Mackus

    HOF

  • Members
  • PipPipPipPipPip
  • 42,640 posts

Posted 11 December 2019 - 02:04 PM

Not a Dan Connolly fan but he wrote an article today heavily implying that the Orioles refuse to leverage teams that want to dump salaries for prospects, like the Cozart trade.

 

*Really* disappointing to hear this.  https://theathletic....al-constraints/

 

Anyone who was "giving the sons the benefit of the doubt", for whatever reasons they chose to ignore the obviousness of this influence in the Machado, Britton, Schoop, and especially Gausman/O'Day trades, they have to be fully aware now that ownership is still very much involved in every decision.  Not only is the budget low, but Elias is not free to use that budget as he sees fit, still needs to be within the framework of the things the Angelos family feels is the right way to operate.  

 

It doesn't really matter who the GM is when the ownership hasn't changed.


  • BSLRobShields, McNulty and BSLMikeRandall like this




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

Partners