Draft Scouting

[ovccsteve]

Anyone else having this issue? I send four guys to scout the upcoming draft. They come back in pairs. Two guys agree on almost everyone, with the other two in total disagreement.

Example: Last pick of first round and I want to draft a college pitcher. Look at top five from each scout. Scout A and Scout B agree on three of five. Scouts C and D give all three of those players a 20 (out of 80), while also agreeing on three of their respective top 5's. Of course, A & B give all those guys 20's. (High school pitchers mirror same thing.)

This is my third draft, and it's happened every time. (Once I only did three, but still no disagreement among the agreeing pair, and I'm assuming if I added a fifth scout, he'd just line up with one of the other pairs--and it would still be a crap shoot which group I should listen to.)

And, to make it more interesting, the two top-rated pitching scouts are in opposite pairs.

[Raidergoo]

I use all my scouts for the draft pool at the earliest possible date. I collect the views on their POT and OVR, dump it into Excel, and start sifting.

It is remarkable how often that consensus is not obtained by the scouting crew. Sometimes they don't seem to be looking at the same guy. I think in the last eMLB draft that I saw 8 (out of 300) solid players in the pool that everyone loved, and the rest of the prospects had different opinions (unless the consensus was that they sucked). Not just a 1/2 difference...they were wildly different opinions.

Nice stuff. Wish it was easier to do, and I wish scouting was more concerned with the evaluation of amateur talent vs. professionals.

[1998 Yankees]

It's not just a matter of how good they are, as indicated by their ratings, but also of scouting preferences; that is, what types of players they prefer. You might see a pattern in this regard of scout pairs favoring the same player attributes.

[ovccsteve]

Originally Posted by 1998 Yankees

It's not just a matter of how good they are, as indicated by their ratings, but also of scouting preferences; that is, what types of players they prefer. You might see a pattern in this regard of scout pairs favoring the same player attributes.

That could be part of it. My problem with it is it's too consistent to be random. IOW, it looks like it's built into the system. I like the uncertainty aspect of using multiple scouts, but this feels almost like obfuscation. (Felt like using a big word.)

[1998 Yankees]

Originally Posted by ovccsteve

I like the uncertainty aspect of using multiple scouts, but this feels almost like obfuscation. (Felt like using a big word.)

Shame on you. Moderators, please do your duty.

[orwell]

Originally Posted by Raidergoo

I use all my scouts for the draft pool at the earliest possible date. I collect the views on their POT and OVR, dump it into Excel, and start sifting.

Can you tell me how you get into Excel? I've always wanted to manipulate and sort draftee ratings but never knew how to get the data.

[injury log]

Originally Posted by ovccsteve

Anyone else having this issue? I send four guys to scout the upcoming draft. They come back in pairs. Two guys agree on almost everyone, with the other two in total disagreement.
[...]

I know exactly what you're talking about. This 'pairing' of scouting reports doesn't seem to have anything to do with scout ratings or scout preferences, from the small number of checks that I've done (and are there any scout preferences for pitchers anyway?). What makes me think it's somehow coded into the game is the alarmingly consistent reports from one scout to his twin; it's not like they agree on a few things, they agree on absolutely everything, and seem to do so throughout their scouting careers. It also seems to happen on other scouting tasks, or at least when I scout my own organization.

Because of this, I don't see any point to have two scout 'twins' both scout the same draft; the second doesn't give me new information. The best prospects are the ones your non-twinned scouts can agree on, and they are, of course, rare after the first round.

[Raidergoo]

Originally Posted by orwell

Can you tell me how you get into Excel? I've always wanted to manipulate and sort draftee ratings but never knew how to get the data.

I click the save to disk button, which produces a HTML page. I copy the data from IE to Excel, apply a little color coding and formatting, and I can swiftly sort through the entire draft in 1 second.

[orwell]

Originally Posted by Raidergoo

I click the save to disk button, which produces a HTML page. I copy the data from IE to Excel, apply a little color coding and formatting, and I can swiftly sort through the entire draft in 1 second.

Thanks. I will try this. I thought that only captured the first screen of data. This would be a huge help when going over the amateur class.

[robster1225]

Make sure you have the option to display scroll bars and not pages. This is found in OOTP preferences. If you have pages on, it will only pull the data from the current page, and you'll have to go through each page and do it. Also, if the amount of players is to large, i.e. inagural draft, then it will use pages no matter what, which really sucks.

[ovccsteve]

From another angle on the draft, does using feeder leagues help, since you have actual stats to use with you two offsetting scouts?

[Hazza]

I had a heart attack about the draft AI when I looked at the results of my sim.

I first noticed that all the international teams were rolling in revenue.

Thats odd, I thought i'd fixed that so it didn't happen.

I then noticed that it was because they were practically selling out each game.

Wierd number two.

How does this have to do with the draft AI? Well, I had set it so the rosters were MLB 25 AAA 25 A 35.

So they were drafting all these players, then having to evaluate who to keep, and then dropping these round 4 guys into the free agent pool after two weeks.

The international teams were picking up these bargain priced players who turned into superstars, bringing the crowds to each game.

I added annother feeder league to reduce the ammount of time this happens. But in theory it should have meant the A teams were full of studs to be, considering the players they were picking up...

[1998 Yankees]

Originally Posted by injury log

I know exactly what you're talking about. This 'pairing' of scouting reports doesn't seem to have anything to do with scout ratings or scout preferences, from the small number of checks that I've done (and are there any scout preferences for pitchers anyway?).

Good point, that. Other than "Favor Pitching" vs. "Favor Hitting" which may make two scouts agree on a set of pitchers over another set of batters, or vice versa, then preferences would seem not to be a factor with pitchers at least.

Maybe it's programmed that a certain range of scouting ratings will rate a certain array of new player potential ratings in a certain way, inevitably leading to matches of similar scouts. Others sets of scouts that share other scouting ratings will respond similarly to other sets of potential ratings, etc.

Originally Posted by ovccsteve

From another angle on the draft, does using feeder leagues help, since you have actual stats to use with you two offsetting scouts?

Certainly! That's why feeder leagues were a great addition to this game, other than the immersion value of following young players' careers from the very beginning.

[injury log]

Originally Posted by 1998 Yankees

Good point, that. Other than "Favor Pitching" vs. "Favor Hitting" which may make two scouts agree on a set of pitchers over another set of batters, or vice versa, then preferences would seem not to be a factor with pitchers at least.

Maybe it's programmed that a certain range of scouting ratings will rate a certain array of new player potential ratings in a certain way, inevitably leading to matches of similar scouts. Others sets of scouts that share other scouting ratings will respond similarly to other sets of potential ratings, etc.

I just checked my scout 'twins'. Their strategy sliders are almost polar opposites; their only agreement is that they both favour OBP over AVG. Their ratings, however, are similar (both are close to max in everything), save for in pitching potential, where there is a 25 point difference, in the /200 scale in the Editor window. So, if I throw caution out the door and draw conclusions from a sample size of one, I'd say that scout 'twinning' is not related to strategy sliders. It may be related to ratings, though in the past I'm almost certain I've seen twinning from scouts with very different ratings.

My suspicion, for which I have absolutely no evidence, but it seems to explain the game behaviour: there is an invisible tag, perhaps a number from 1 to 8 or 1 to 16, assigned to each scout, and correspondingly a tag for each player (or at least each draft-eligible player), which causes scouts to overrate players with matching tags. Two scouts with identical tags would then give consistent scouting reports; they'd overrate the same guys. It may well be a design feature to ensure that there are still a few well-rated (overrated, really) prospects deep into each draft, and to ensure that a scout remains somewhat consistent from evaluation to evaluation. If this rampant speculation approximates how the game works, it seems like a good game design strategy to me; my only issue is the scout pairing, which superficially seems arbitrary, i.e. doesn't seem explicable based on anything we can see in-game.

[ovccsteve]

Originally Posted by injury log

I'm almost certain I've seen twinning from scouts with very different ratings.

I don't have a problem when different rated scouts agree; it's when like-rated scouts disagree. I send out two guys with 19's or 20's for rating pitchers' potential, which should mean I can trust them more than a guy with a 14. Unfortunately, they're just as likely to disagree totally, which means, best I can tell, when it comes to the draft, I'd be just as well served signing three poor scouts to Rookie Ball, promote them to the Majors with their cheap contracts, and let them scout the draft, saving me about $300K a year I could use for salary or profit.

The twinning really bothers me. If they all disagreed along a spectrum, with the better rated scouts doing a better--but still not perfect--job, that would be okay. But the twinning is just, to use my big word again, obfuscation n the game design.

[orwell]

Originally Posted by robster1225

Make sure you have the option to display scroll bars and not pages. This is found in OOTP preferences. If you have pages on, it will only pull the data from the current page, and you'll have to go through each page and do it. Also, if the amount of players is to large, i.e. inagural draft, then it will use pages no matter what, which really sucks.

I want to thank you for that tip. Even though it was right there in the options, I had never noticed it and was always annoyed that you could not get more than one page in the "write to disk" command. This has been a huge help with the way I scout. I create a spreadsheet now and average the ratings of all five of my scouts, and sort through the positions that way.

[Curtis]

Originally Posted by Raidergoo

I use all my scouts for the draft pool at the earliest possible date.

So do I. After I get all six reports back, I make a list of everyone who's named on each scout's top twenty list. Each guy gets one point for each scout who mentions him, regardless of whether the scout ranks him 20 or 1, so a draftee named by everyone swcores a '6' and someone only named once gets a '1'.

When I draft, I choose among my sixes until they're gone, then among the fives, fours, etc. I only skip if I'm getting overloaded at a position.

I've never seen this 'twinning'. I usually have four scouts who more or less agree, then two outliers to keep things interesting.

[koohead]

I must be missing something, but how do you send your scouts to scout the draft pool??

[denisonh]

Good question

Originally Posted by koohead

I must be missing something, but how do you send your scouts to scout the draft pool??

[denisonh]

Aaaaahh, you have to go to the option "review the draft pool" from the managers' home screen.

Originally Posted by denisonh

Good question