Value Based Drafting in Baseball - Fantasy Baseball Cafe 2015 Fantasy Baseball Cafe
100% Deposit Bonus for Cafe Members!

Return to Baseball Leftovers

Value Based Drafting in Baseball

Moderator: Baseball Moderators

Re: Value Based Drafting in Baseball

Postby OaktownSteve » Wed Feb 27, 2013 2:02 pm

Izenhart wrote:
J35J wrote:You're thinking too much. This game has way to much luck involved to take it to the extremes you may be wanting to.



Well, no one is trying to come up with one be all list for the single best ADP here. They are simply stating the best pick on the board is always changing depending on your team composition and player availability. To come up with a best case scenario draft where every team makes the best pick in every round is simply not possible. All we are discussing is what may be the best strategy when trying to give yourself the best shot at winning, without relying on the big fluffy luck dragon.

Value based drafting works best when value meets tier meets need. Setting yourself up for these picks can be like playing chess and thinking 2-3 moves or rounds ahead. I think that is the best way to construct a team. Knowing the ADP of players and also your opponents strategies and favorites will help you out immensely here.


I like your "value meets tier meets need" and also the chess analogy. I alluded to a super computer that was powerful enough to run every possible draft scenario and pinpoint the optimal selection for each selection in the draft (probably not do-able) but I think back to Big Blue vs Kasparov in the chess world and apply it to a fantasy draft.

Big Blue competed by running millions of scenarios as many moves out as possible with some logic built in but mostly relying on blunt force. Kasparov countered that blunt force computing power by narrowing the number of lines that he found viable based on experience and the superior pattern recognition in the human brain versus the computer chip.

In fantasy, like a chess expert, the more "games" you play, the more you become familiar with likely outcomes based on how you draft. You assemble your tools (values, tiers, cheat sheet, etc) so that you are better able to recognize where value is meeting tier is meeting need in your draft.

And while luck is a factor that prevents the best players from winning every time, the better players win more often and minimize the impact of luck through careful roster construction and in season management. If it was all luck, where’s the fun in that.
OaktownSteve
Little League Legend
Little League Legend


Posts: 22
Joined: 26 Feb 2013
Home Cafe: Baseball

Re: Value Based Drafting in Baseball

Postby Izenhart » Wed Feb 27, 2013 3:11 pm

Yes, I was an avid chess fan at one time and loved to memorize the more unorthodox openings, believing that setting yourself up for the endgame and having a strategy on how to get there was important when deciding which route to take. Fantasy baseball is a lot like that to me in snake drafts, I build my team backwards and fill in the blanks as I go, setting late round targets first and having a backup plan for each failed pick.

This year my mock drafts are much stonger when I get a 1-3 pick. It's a bigger advantage this year than any in recent memory to have a top pick.
Izenhart
Major League Manager
Major League Manager

User avatar

Posts: 1611
(Past Year: 286)
Joined: 22 Mar 2012
Home Cafe: Baseball
Location: Big Papiville

Re: Value Based Drafting in Baseball

Postby OaktownSteve » Wed Feb 27, 2013 3:19 pm

Izenhart wrote:Yes, I was an avid chess fan at one time and loved to memorize the more unorthodox openings, believing that setting yourself up for the endgame and having a strategy on how to get there was important when deciding which route to take. Fantasy baseball is a lot like that to me in snake drafts, I build my team backwards and fill in the blanks as I go, setting late round targets first and having a backup plan for each failed pick.

This year my mock drafts are much stonger when I get a 1-3 pick. It's a bigger advantage this year than any in recent memory to have a top pick.


I like that idea of reverse engineering from the end game. I have found that the most successful players in the most competetive leagues are consistently finding max value at the end of the draft. Everybody talks about who goes in what order in the first round of the draft but it generally is not where the game is won or lost. Much like chess opennings, the first few picks dictate where you are going strategically but not necessarily tactically later on.
OaktownSteve
Little League Legend
Little League Legend


Posts: 22
Joined: 26 Feb 2013
Home Cafe: Baseball

Re: Value Based Drafting in Baseball

Postby J35J » Wed Feb 27, 2013 3:21 pm

OaktownSteve wrote:And while luck is a factor that prevents the best players from winning every time, the better players win more often and minimize the impact of luck through careful roster construction and in season management. If it was all luck, where’s the fun in that.

The better the league the more luck plays a role....add in that "in season management" is likely more important than the draft as well. Ender has pretty much said it right. Don't get bogged down with precisely what specific order each and every player needs to be in, because it changes based on the flow of the draft. Throw em in tiers and work the draft with your knowledge along with that.



OaktownSteve wrote:The main takeaway here is that a draft is a really, really complex problem that can't be solved by tiers or values.

And this is why just throwing them in tiers and letting your knowledge lead the way is the way to go.



Actually, I think we are all in agreement here. The draft is a flowing beast that changes...you can't come up with a perfect draft sheet. Your thinking about each player has to be dynamic right along with the draft.
J35J
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicFantasy ExpertCafe WriterCafe RankerMock(ing) DrafterPick 3 Weekly Winner
Posts: 10467
(Past Year: 361)
Joined: 26 Jan 2004
Home Cafe: Baseball

Re: Value Based Drafting in Baseball

Postby Izenhart » Wed Feb 27, 2013 3:36 pm

OaktownSteve wrote:
Izenhart wrote:Yes, I was an avid chess fan at one time and loved to memorize the more unorthodox openings, believing that setting yourself up for the endgame and having a strategy on how to get there was important when deciding which route to take. Fantasy baseball is a lot like that to me in snake drafts, I build my team backwards and fill in the blanks as I go, setting late round targets first and having a backup plan for each failed pick.

This year my mock drafts are much stonger when I get a 1-3 pick. It's a bigger advantage this year than any in recent memory to have a top pick.


I like that idea of reverse engineering from the end game. I have found that the most successful players in the most competetive leagues are consistently finding max value at the end of the draft. Everybody talks about who goes in what order in the first round of the draft but it generally is not where the game is won or lost. Much like chess opennings, the first few picks dictate where you are going strategically but not necessarily tactically later on.


Last year my saving grace was Chris Sale. This year I'm not so confidant I have someone going way later than they should, so I'm trying more to minimize risk rather than putting all my eggs in the maximize potential basket.
Izenhart
Major League Manager
Major League Manager

User avatar

Posts: 1611
(Past Year: 286)
Joined: 22 Mar 2012
Home Cafe: Baseball
Location: Big Papiville

Re: Value Based Drafting in Baseball

Postby buffalobillsrul2002 » Fri Mar 01, 2013 2:33 am

You know, there's probably a model you could actually create that would work. It's just gotta take a few extra things into account:

1) Multi-positional flexibility - there are many guys who have dual-elgibility. This is probably the simplest one to solve- I think you just put everyone in at one position - their "best" position - I think this is pretty easy to solve - just set up all your projections beforehand and then slot these guys in (in order from "best stats" to "worst stats" at their positions. But it still throws a wrench into the game.

2) Ability to strategize within your roster - I think this is highly overlooked - in football and basketball, it's really hard (no matter how much people try) to figure out how you are going to manage week-to-week until you get into the middle of the season, since so much of those two games are based solely off opportunity. Because everyone gets the same opportunity in baseball, it's often easier (especially with pitching) to figure out how approximately how often you are going to sit/start a guy. I would say this gets thrown into your analysis as a variable (for pitching is easy, you only project stats for when you think you'll play them). For hitting, I have no idea what you do.

3) What the heck is replacement-level player? For this, I think you have to do two things. First, you won't be able to calculate one replacement-level player; you're going to have to look at a whole bunch of guys who are in the replacement range and then average their stats. The next problem comes in that you are going to look at the distribution of starters for a give statistic. I would argue that if there is one first baseman who steals 40 bases and the rest steal 5-10, then those 35 extra steals have more value then a situation where there are say, 10 SS who steal 30 bases but then nobody around the replacement level does. I'm not really sure how one accounts for this well.

4) Different stats have different amounts of projectability/variation. I don't know how one accounts for this though precisely, i think you downgrade some of the more projectable stats a little bit. But you can't downgrade too much or you'll end up with a one-dimensional team.

I think you could create some sort of VBD model, especially in points. Roto you're going to have to add quite a bit of "feel" into the model to get to a correct solution.
buffalobillsrul2002
Major League Manager
Major League Manager


Posts: 1349
(Past Year: 3)
Joined: 17 Jun 2004
Home Cafe: Baseball

Re: Value Based Drafting in Baseball

Postby TheRock » Sun Mar 03, 2013 3:52 am

I had a look at your spreadsheet, I think you're on the right track on some things. Determining how to value players gets discussed here now and then, do a search and you'll probably find a lot of threads on the topic. Here's an article you can start with, I think it will answer some of the questions you're stuck with (and some I don't think you know you are yet ;-) ) Like how to resolve that a closer with great ratios is not worth as much to you as a starter with great ratios. And that 10 additional homeruns is worth more than 10 additional rbis.

You've started this exercise with the notion of applying what you know VBD from fantasy football to fantasy baseball. It actually has already been here, you'll here people talk about positional scarcity, or replacement level players, and that's generally what they're talking about. Where I would challenge you in particular is the assumption that you need to. Now catcher is usually the one exception, but this year I am making no adjustments in my player values for catchers. Two catcher leagues - different story. But this isn't football where quarterbacks generally outscore every other position and tight ends don't do crap except for Gronkowski. Give some thought to whether two identical players should be valued differently if one plays outfield and ones plays shortstop and why. What if your league has a MI spot, 5 OF spots, and 2 Util spots, how does that change things? There's not really a consensus, but think it through before you jump right into compiling a spreadsheet for something you might ultimately decide you don't need to pursue.
Image
TheRock
General Manager
General Manager

User avatar
CafeholicCafe WriterMock(ing) Drafter
Posts: 3055
(Past Year: 6)
Joined: 16 Apr 2003
Home Cafe: Baseball
Location: America's Heartland

Re: Value Based Drafting in Baseball

Postby Ender » Tue Mar 05, 2013 10:35 am

I can't stress enough that whatever you use for replacement level at a position is wrong, especially if you are basing it off of a projection system. Projections regress everything heavily especially for unknown commodities which means their lower level rankings are always pessimistic compared to reality. If you have to come up with a replacement level player you are much better off using last years stats and looking for replacement from those stats which are more realistic than any projection system. Last year 3B was really shallow going into the draft and then like 6 different 3B broke out making it a deep position in reality. Going into last year C looked shallow in a 2 C league and then AJ had his year, Rosario broke out, Lucroy became very viable, Ruiz had a career year etc and it became very deep. There just is no way to know what replacement level is and trying to adjust your draft board based on some phantom replacement level is just not a very good idea.

You also really need to account for the fact that players get shifted away from strong positions and to scarce positions. Seager isn't a 3B he is a 2B. Prado is a 2B. Zobrist is a SS. You have 12-14 players that are most likely CI that are being put into UT slots. Hart is an OF etc. Once you account for all of that you completely dilute the deep positions and remove most of the positional scarcity.

The other issue is this ignores risk. The risk curve more or less is an inverse of what people view as the positional scarcity curve. RP is the riskiest position since once you lose saves you lose most of your value. C is probably the next riskiest, followed by SP since the stats are all so erratic. After that you have SS/2B as the riskiest followed by 3B/CF since they get hurt more and finally the most reliable positions are corner OF and 1B. Drafting for positional scarcity is nothing but a trap.
Ender
Hall of Fame Hero
Hall of Fame Hero

CafeholicFantasy Expert
Posts: 7733
Joined: 30 Apr 2004
Home Cafe: Baseball

Previous

Return to Baseball Leftovers

Who is online

Users browsing this forum: No registered users and 13 guests

Forums Articles & Tips Sleepers Rankings Leagues


Get Ready...
The 2015 MLB season starts in 21:28 hours
(and 96 days)

  • Fantasy Baseball
  • Article Submissions
  • Privacy Statement
  • Site Survey 
  • Contact