31 replies to this topic

[Luke]

When ranking prospects... something that is admittedly an artificial construct of limited actual utility... what do you give more weight to: a player's potential, or the likelihood of that player reaching the majors.

Let's set up a hypothetical to make things easy. Each of these five fictional players are being rated in two categories on a fairly standard 2 - 8 scale (analogous to the 20 - 80 scale you're used to). "Potential" refers to how good a player is likely to be when he reaches the majors. This is NOT the same thing as Ceiling. I'm not talking about how good he can be in a near ideal scenario, but how good he is likely to be. I know it's vague, but this is a hypothetical. Just go with it for now.

"Likelihood" refers to the chance that the player will make it to the majors and stick around for awhile. Again, vague, go with it for now.

Albert Potential: 5 Likelihood: 4
Barry Potential: 4 Likelihood: 7
Carl Potential: 8 Likelihood: 3
David Potential: 6: Likelihood: 5
Edgar Potential: 5 Likelihood: 6

In what order would you rank these five players?

Edgar is easily ranked over Albert (both are P:5, but Edger has a higher L). After that, it gets more interesting.

[calicubsfan007]

Umm... Let me see...
1.Barry
2.Carl
3.David
4.Edgar
5.Albert
I am sure I am wrong on the order, but at least I can say that I have tried.

[Luke]

I'm not sure there is a wrong order. I'm not sure there is a right one, either.

[calicubsfan007]

Oh, I get it. This kinda shows how high we put potential and likelihood. Shows what our general view of prospects is, right?

[MichiganGoat]

DICK TIDROW #1

[Spencer]

Edgar
Carl
David
Barry
Albert

[Luke]

I'm genuinely curious what people think.

It comes up because I've been working on my prospect rankings, and I've been trying to nail down an algorithm I can use to at least block out my rankings, and I've been getting some widely variant results based, primarily, around very slight changes in how I weight a couple of factors that can be broadly summarized (if poorly) as Potential and Likelihood.

This thread is exactly what it says it is... a philosophical question. How do you value these two factors on a comparative basis.

[scorecardpaul]

Ok, I'll jump off that bridge....
I think it would be safe (at quick glance) to say your original list is backwards.


Albert Potential: 5 Likelihood: 4
Barry Potential: 4 Likelihood: 7
Carl Potential: 8 Likelihood: 3
David Potential: 6: Likelihood: 5
Edgar Potential: 5 Likelihood: 6

Edgar is most valuable because there should be a pretty good chance of producing an average player
David is next because you have a decent chance to produce a good player
Carl is the intriging one for me, not very likely to succed, but if he does, you have a star. you keep Carl around for a long time to find out
Barry is next, because he is pretty sure to make it to the show. at least as a cheap role player.
Albert is last because not much of a chance to succede even as average.

I can understand why this would be so hard to predict a prospect.

now for fun lets put some real names in place of the generic ones??
I am assuming we have a guy like Carl playing in Venezualia right now???

[scorecardpaul]

is this way off the wall for a made up comparison??


Albert Potential: 5 Likelihood: 4
Barry Potential: 4 Likelihood: 7 Logan Watkins
Carl Potential: 8 Likelihood: 3 a pitcher who throws 100 mph/wild
David Potential: 6: Likelihood: 5 Soler
Edgar Potential: 5 Likelihood: 6 Almora
7 5 Bajez

sorry if this is stupid

[calicubsfan007]

It is a logical comparison. And Happy Bday.

[OCCubFan]

Because both factors are important, the simplest (not necessarily best) algorithm is to multiply the two factors. Then,

Albert Potential: 5 Likelihood: 4 --> 20
Barry Potential: 4 Likelihood: 7 --> 28
Carl Potential: 8 Likelihood: 3 --> 24
David Potential: 6: Likelihood: 5 --> 30
Edgar Potential: 5 Likelihood: 6 --> 30

Therefore the ranking is: Edgar & David, Barry, Carl, Albert.
If this ranking doesn't seem correct to you, figure out why. Perhaps you think Potential should be rated higher; if so, your algorithm might be: (Likelihood) times (Potential squared). In that case, the rankings become:

Albert Potential: 5 Likelihood: 4 --> 100
Barry Potential: 4 Likelihood: 7 --> 112
Carl Potential: 8 Likelihood: 3 --> 192
David Potential: 6: Likelihood: 5 --> 180
Edgar Potential: 5 Likelihood: 6 --> 150

That is: Carl, David, Edgar, Barry, Albert.

[OCCubFan]

Both factors are important in that Potential with no Liklihood of reaching the majors is useless (and unlikely), and high Liklihood with no Potential (a seeming contradiction). Because the rating scale is 2 to 8, 2 corresponds to Zero Potential or Liklihood. Therefore, I wish to revise my previous algorithms to multiplying (Potential - 2) by (Liklihood - 2). Then we have

Albert Potential: 5 Likelihood: 4 --> Score = 6
Barry Potential: 4 Likelihood: 7 --> Score = 10
Carl Potential: 8 Likelihood: 3 --> Score = 6
David Potential: 6: Likelihood: 5 --> Score = 12
Edgar Potential: 5 Likelihood: 6 --> Score = 12

Therefore the ranking is: Edgar & David, Barry, Carl & Albert.
If this ranking doesn't seem correct to you, figure out why. Perhaps you think Potential should be rated higher; if so, your algorithm might be: (Likelihood - 2) times [(Potential -2) squared]. In that case, the rankings become:

Albert Potential: 5 Likelihood: 4 --> Score = 18
Barry Potential: 4 Likelihood: 7 --> Score = 20
Carl Potential: 8 Likelihood: 3 --> Score = 36
David Potential: 6: Likelihood: 5 --> Score = 48
Edgar Potential: 5 Likelihood: 6 --> Score = 36

That is: David, Carl & Edgar, Barry, Albert.

[Cubbie Blues]

David Potential: 6: Likelihood: 5
Carl Potential: 8 Likelihood: 3
Edgar Potential: 5 Likelihood: 6
Barry Potential: 4 Likelihood: 7
Albert Potential: 5 Likelihood: 4

It would also matter to me what level they are all at. If Carl is a 3 and in AA or AAA I would drop him down a slot maybe even two. As it is without knowing what level they are at I have Carl and Edgar very close anyway.

Edit:
Seriously, OCC, I didn't see your ranking above me when I listed mine. It was just the way I saw the numbers sorting out in my head.

Edited by Cubbie Blues, 01 February 2013 - 07:11 AM.

[Spriggs]

..... Perhaps you think Potential should be rated higher; if so, your algorithm might be: (Likelihood - 2) times [(Potential -2) squared]. In that case, the rankings become:

Albert Potential: 5 Likelihood: 4 --> Score = 18
Barry Potential: 4 Likelihood: 7 --> Score = 20
Carl Potential: 8 Likelihood: 3 --> Score = 36
David Potential: 6: Likelihood: 5 --> Score = 48
Edgar Potential: 5 Likelihood: 6 --> Score = 36

That is: David, Carl & Edgar, Barry, Albert.

That is pretty much exactly how I went about it. Potential does mean a bit more to me than probably most, so I kept jacking with the numbers until Carl was first. For me, it's:

Carl
David
Edgar
Barry
Albert

[SirCub]

I think you'd have to give the edge to the player with slightly more potential over one with slightly more likelihood (or vastly more potential over vastly more likelihood). For my money, I'd go:

Carl
David
Edgar
Barry
Albert