Auction Value generators

[bryanbutler]

You still can't beat the pen, paper, highlighter method IMHO.

ugh. i disagree. i find it *much* faster to search in a file on the computer than to do it with my eyes on a piece of paper. and i continually update my values (re-run my own value generator) as the auction proceeds.

[JudeBaldo]

ugh. i disagree. i find it *much* faster to search in a file on the computer than to do it with my eyes on a piece of paper. and i continually update my values (re-run my own value generator) as the auction proceeds.

That's interesting. What I do, and I learned this from TheDane, is to essentially take all of the digital files and crystalize them into a few pieces of paper with positional tiers. I figure, if you've done the previous research, you don't need all of the tech with you. I find that if I'm relying too much on the digital files, I'm missing out on the ebb and flow of the draft.

Do you run one or two monitors?

[Kevin Seitzer]

That's interesting. What I do, and I learned this from TheDane, is to essentially take all of the digital files and crystalize them into a few pieces of paper with positional tiers. I figure, if you've done the previous research, you don't need all of the tech with you. I find that if I'm relying too much on the digital files, I'm missing out on the ebb and flow of the draft.

That was always my method, too. I was able to connect much more with what was going on in the auction if I was crossing out names on a piece of paper rather than trying to follow a spreadsheet or something else on a computer monitor.

[The Dane]

I'm all low-tech at a live draft, and I have never felt in any instance that any owner was ever served by going high-tech. I tend to be very competitive in my live auctions (11 titles over three leagues) and I think it's mainly because I do not bog myself down with useless details when I'm trying to OUTTHINK, to ANALYZE, to ASSESS. The fact that so-and-so projects 15 SBs and some other so-and-so projects 25 SBs and some homer thinks 45 SBs but that homer was right about some other thing at some other time... screw all that. I like this player and I need SBs, and I can clearly and quickly see how many SBs are left on the table, so I get him. This frees me up to know what everyone is doing, seeing what they need, price-enforcing, screwing up other owners' drafts as much as I can. That poor schmoe locked behind his keyboard is married to his tech and he can't think as clearly as I can. I will beat that guy 8 times out of 10, because he has put himself at a extreme disadvantage. Most of the information even the good programs provide is just noise that doesn't really help and is more of a distraction from the things that are really important.

I read everything I can from the day I get the Forecaster until the moment I get in the car to head for the draft. If another owner knows something useful about the player pool that I don't, all the more power to him.

I bring one piece of paper and one pen. The paper has available players listed by position in tiers with rough dollar values, that are there really just for sorting purposes. It usually goes considerably deeper than the league will need and contains no other information, no stats, no notes, and usually not even full first names. Sometimes if a league has a rule that a player MUST be on a 25-man or 40-man or make the team out of ST sort of rule, I will sometimes put a little dot next to players that qualify.

Two caveats:
1) I do use a spreadsheet draft-tracker thingee. It computes money left, spent and highest bid, and I can see who needs what. I do it for the math, basically, and if someone else did it and projected it, I would not need it.

2) This does not necessarily apply to online auctions. I suppose Rotolab may actually be useful there, but I'd have trouble managing that AND the CBS draft room. I've had much less success in those leagues (only 3 titles over two leagues), because I think my best advantage -thinking clearly at the draft- is muted by the nature of onlineness.

[bryanbutler]

i see your point in a live auction - i was speaking to online ones (i haven't done a live auction in 15 years!). but you seem to be implying that doing it with a pen and piece of paper is faster than doing it digitally. at least i think that's what your argument boils down to. make things simple and fast so that you can pay attention to the dynamics of the owners (body language, vocal tone, etc.), and have some sort of feel for when GMs are overpaying or underpaying, runs on positions, etc.

i disagree with the fundamental premise that pen and paper is faster than digital. i have three text files i use during an auction - one tracks all players taken; one lists current estimated value, which i calculate with my own program using VORP (with the RP stats varying by position); one lists remaining players by position. actually there is a fourth file, that tracks current estimated stats for each of the GMs, so that i know who needs what, but i don't refer to that except for every round or two. all my auctions are in CBS now, so i use that to track what each GM needs positionally. i run a simple text editor (vi), in a multi-tabbed terminal window, so can find a player in any one of these three of my own files in seconds (click the right tab, hit the '/' key and type in enough of the player's name to be unique, hit return).

in that vein, one thing you say confuses me. you say "...I can clearly and quickly see how many SBs are left on the table..." this is where pen and paper seems to be worse than digital to me. you've got to scan your entire piece of paper, by eye, pick out the players who might get a decent number of SBs, and remember them, total them up in your head, compare that to what GMs have SBs and what GMs need them, and then make your bids. maybe if you have an auction that has long allowed periods between bids that might work. seems to me it'd be too slow. for me, i simply sort the file of stats of remaining players (using the command line "sort" command in another terminal window), and voila, i've got a sorted list of SBs still on the table. only want middle infielders who will get you SB? pipe the results of the command line "sort" into a "grep", and voila, sorted SBs for middle infielders.

now, i agree that if somebody's tech is slow - if it's taking them too much time to type things into spreadsheets, etc., then it's a failure. but, again, i can do all the things i do on my computer in seconds.

i never take predictions of either stats or value too seriously - you can get yourself in real trouble doing that. so your assumption that somebody that is using a computer is "married to his tech" is not true for everybody.

roto is a game of stats. if my "stats" are "better" than the GM with the pen and paper going into and during the auction (and this is not just raw numbers, but understanding the players and the league GMs, and this includes understanding the uncertainties), then i will beat that guy 8 times out of 10, unless i do something stupid.

[cavebird]

I'm all low-tech at a live draft, and I have never felt in any instance that any owner was ever served by going high-tech. I tend to be very competitive in my live auctions (11 titles over three leagues) and I think it's mainly because I do not bog myself down with useless details when I'm trying to OUTTHINK, to ANALYZE, to ASSESS. The fact that so-and-so projects 15 SBs and some other so-and-so projects 25 SBs and some homer thinks 45 SBs but that homer was right about some other thing at some other time... screw all that. I like this player and I need SBs, and I can clearly and quickly see how many SBs are left on the table, so I get him. This frees me up to know what everyone is doing, seeing what they need, price-enforcing, screwing up other owners' drafts as much as I can. That poor schmoe locked behind his keyboard is married to his tech and he can't think as clearly as I can. I will beat that guy 8 times out of 10, because he has put himself at a extreme disadvantage. Most of the information even the good programs provide is just noise that doesn't really help and is more of a distraction from the things that are really important.

I read everything I can from the day I get the Forecaster until the moment I get in the car to head for the draft. If another owner knows something useful about the player pool that I don't, all the more power to him.

I bring one piece of paper and one pen. The paper has available players listed by position in tiers with rough dollar values, that are there really just for sorting purposes. It usually goes considerably deeper than the league will need and contains no other information, no stats, no notes, and usually not even full first names. Sometimes if a league has a rule that a player MUST be on a 25-man or 40-man or make the team out of ST sort of rule, I will sometimes put a little dot next to players that qualify.

Two caveats:
1) I do use a spreadsheet draft-tracker thingee. It computes money left, spent and highest bid, and I can see who needs what. I do it for the math, basically, and if someone else did it and projected it, I would not need it.

2) This does not necessarily apply to online auctions. I suppose Rotolab may actually be useful there, but I'd have trouble managing that AND the CBS draft room. I've had much less success in those leagues (only 3 titles over two leagues), because I think my best advantage -thinking clearly at the draft- is muted by the nature of onlineness.

Please tell me you use the old RJ spreadsheet. That's what I do. That thing and print outs of my values with pen and paper. I almost want to buy RotoLab to see their values because 75% of my league uses it.

[mjl]

pipe the results of the command line "sort" into a "grep"

NNNNERRRRRRD

and that is intentionally in all caps so stop auto-editing me you stupid web site

[JudeBaldo]

That was always my method, too. I was able to connect much more with what was going on in the auction if I was crossing out names on a piece of paper rather than trying to follow a spreadsheet or something else on a computer monitor.

Exactly.

There may be a scientific basis for it.

http://lifehacker.com/5738093/why-you-learn-more-effectively-by-writing-than-typing

Also, we know from study after study that people that single task are superior to people that multi-task (doing the single task condition with the addition of another task) when it comes to response rates. Of course there are always exceptions.

Does anyone else notate their reading material? For example, I'll be marking up the Starting Pitching Guide like crazy in the next few weeks.

[The Dane]

i have three text files i use during an auction - one tracks all players taken; one lists current estimated value, which i calculate with my own program using VORP (with the RP stats varying by position); one lists remaining players by position. actually there is a fourth file, that tracks current estimated stats for each of the GMs, so that i know who needs what, but i don't refer to that except for every round or two. all my auctions are in CBS now, so i use that to track what each GM needs positionally. i run a simple text editor (vi), in a multi-tabbed terminal window, so can find a player in any one of these three of my own files in seconds (click the right tab, hit the '/' key and type in enough of the player's name to be unique, hit return).

in that vein, one thing you say confuses me. you say "...I can clearly and quickly see how many SBs are left on the table..." this is where pen and paper seems to be worse than digital to me. you've got to scan your entire piece of paper, by eye, pick out the players who might get a decent number of SBs, and remember them, total them up in your head, compare that to what GMs have SBs and what GMs need them, and then make your bids. maybe if you have an auction that has long allowed periods between bids that might work. seems to me it'd be too slow. for me, i simply sort the file of stats of remaining players (using the command line "sort" command in another terminal window), and voila, i've got a sorted list of SBs still on the table. only want middle infielders who will get you SB? pipe the results of the command line "sort" into a "grep", and voila, sorted SBs for middle infielders.

This isn't rocket science. The numerical detail that your multiple sheets provide is 99% useless in a live auction, in my opinion. Projections are rarely so accurate that your using "accurately" actually helps. Here's my point. Here are my hitting keepers in my NL only. Let's pretend I'm drafting from this list and I'm thinking about SBs.

C-
C-
1- PGoldschmidt
2-
3-
S- JRutledge
CI- RHoward
MI- SCastro
OF- RBraun
OF- GStanton
OF- AEaton
OF-
OF-
UT-

I know offhand that it takes roughly 200 SBs to win my league in SBs. I'm shooting for slightly less than that. I can round very quickly that I have roughly 80 SBs currently on my team, 10 from Goldy, 15 from Eaton, 10 from Rutledge, 40 from Castro/Braun and 5 from Stanton/Howard. So, in my head, I'm thinking I need about 100 SBs from this draft. As I get a player, I just check it off in my mind. Do I have to be accurate? No. The projections aren't accurate anyway. If I get to 150 or so, I'm cool. I'll just trade or FAAB from that point. And I do the same with HR, R and RBI. Same with pitching.

I don't NEED to be accurate. Accuracy is a myth. None of that information your sheets generate helps me at all. It's all pointless. Now, if I was drafting my team using 2013 records, then a sheet like that would be VERY useful because I could clearly know exactly how much I have and how much the other owners have and I could draft a winner with perfect knowledge. Really. Look at last year's draft sheets. Many of your assumptions look damn foolish right now. ("Psh. I like Trout as much as anyone but there is NO WAY I am going to double digits for a rookie who didn't even make the team out of Spring Training." "Go to $2 on Everth Cabrara!? No way. He doesn't even have a regular gig.") What were perfectly reasonable guesses a year ago are totally laughable absurdities now. So, think of all the TIME spent collating, grepping, and analyzing essentially flawed data. I don't bother with it. I know my player pool. I know my fellow owners. I draft the team I want. If I lose, it is always because I valued the wrong players too highly or too lowly, and not because someone's software outdrafted me.

Again, this may be different in an online auction. I do not have the same success there with this methodology. My showings in RJEL (and JWRL before it) are perfect examples. However, I'm not entirely sure that it is because it is online or that it's mixed and I am still a neophyte with the AL. My NL Icestorm drafts are online and I have three titles there.

But, hey man, do what works for you. I'm a minimalist.

For what it's worth, when I spoke to Ron Shandler about this, he said he brings just what I bring, and I think this is somewhat common among industry "experts." Not that that is proof of anything; just saying.

[bryanbutler]

NNNNERRRRRRD

and that is intentionally in all caps so stop auto-editing me you stupid web site

heh.

but this is kinda my point. for those of us that can do that kind of thing without thinking about it, it's second nature, and fast.

[bryanbutler]

This isn't rocket science.

no, it's just a bit of statistics.

The numerical detail that your multiple sheets provide is 99% useless in a live auction, in my opinion.

i think you've missed my point. perhaps i didn't make it very well. it's not the *accuracy* on a specific projection or number that's important, it's the ability to access it quickly and incorporate it into what you should do. so your lengthy description of how to get your SBs for your team? i assert that it's just as easy to do that with a "digital" technique as with your "check it off in my mind" one.

you must have been exposed to some oddball tech-savvy roto folks in your auctions. tech-savvy (or "digital", or "nerd" if you will ) doesn't mean you buy into the numbers precisely. or it shouldn't. you just use the tools to gather the information you want or need quickly.

I don't NEED to be accurate. Accuracy is a myth.

again, you're equating using provided information digitally to assuming it's precise (or accurate) - which i don't do.

But, hey man, do what works for you. I'm a minimalist.

to this i agree. if you can make pen and paper and your pre-draft research (and memory of it) work for you, that's great.

For what it's worth, when I spoke to Ron Shandler about this, he said he brings just what I bring, and I think this is somewhat common among industry "experts." Not that that is proof of anything; just saying.

to each their own. lots of older folks are uncomfortable with technology. or unable to do things quickly with it. for those of us who aren't, and can, i think it's a huge benefit.

[JudeBaldo]

no, it's just a bit of statistics.



i think you've missed my point. perhaps i didn't make it very well. it's not the *accuracy* on a specific projection or number that's important, it's the ability to access it quickly and incorporate it into what you should do. so your lengthy description of how to get your SBs for your team? i assert that it's just as easy to do that with a "digital" technique as with your "check it off in my mind" one.

you must have been exposed to some oddball tech-savvy roto folks in your auctions. tech-savvy (or "digital", or "nerd" if you will ) doesn't mean you buy into the numbers precisely. or it shouldn't. you just use the tools to gather the information you want or need quickly.



again, you're equating using provided information digitally to assuming it's precise (or accurate) - which i don't do.



to this i agree. if you can make pen and paper and your pre-draft research (and memory of it) work for you, that's great.



to each their own. lots of older folks are uncomfortable with technology. or unable to do things quickly with it. for those of us who aren't, and can, i think it's a huge benefit.

Bryan, I'm curious, can you give me an example when you'd need to look for data?

[bryanbutler]

Bryan, I'm curious, can you give me an example when you'd need to look for data?

depends on what you mean by "data." for example, i rarely look at projected stats for a particular player. i do bulk queries like what i described above (give me all remaining 2B and sort them on this item), but i don't look at a player's projected stats right when he's nominated, because i know at least roughly what they'll be. well, the exception is when we get to dollar days - i often don't know those guys so i'll look them up.

another example is that i think it behooves you to know, as the auction progresses, where all of the GMs are in the stats categories. now, you might be able to roughly track what you've got on your own team, but i doubt you'll be able to do it for every other team, with a pen and paper and just your brain. having all the data there at your fingertips allows you to know, almost instantly, what the other teams have and are looking for. again, take it with a grain of salt (you need to have a feel for the uncertainties), but if your projections show large differences in a category total for a GM, they're probably right. so, if a speed guy comes up, and you know a particular GM has a need for speed (and especially if it's at a particular position that doesn't have much left) because you can type a single command into a terminal and get the full league category totals up to this point in the auction, even if you don't need that player you can feel pretty safe bidding the other GM up a bit. or, if you do need that player, you might have a bidding war on your hands and it's good to know that at the start (a pre-empt might work, for instance). if you have a savant's ability to remember things you might be able to do all of that in your head, but for most of us mere mortals it's impossible.

[cool hand]

i'm bumping this cause i think we got a bit off track with the thread. personally - i come to the draft with two sheets of paper, albeit on very small font. one sheet with with every offensive player listed by positional category, ranked by a composite of values of my choosing (LABR & TOUT prices usually make it into my exceedingly simple algorithm) to get a feel for relative values of players, not so much hardened values for forecasting draft salaries. the other sheet is a similar ranking of pitchers. both sheets include kept players. players on each sheet are highlighted in various colors representing a number of categories, adhering to a code only known by yours truly.

as far as automated drafting and computer programs - our league has been drafting for the last 14 years out of a las vegas sports book, and we're not allowed to have laptops in play.

BUT - GETTING BACK TO THE ORIGINAL POST
it seems as though we've not yet come up with a (free) software or website that will calculate auction values including a given set of parameters including keepers (ie, incorporating inflation into the results)? it sounded like we were close with the lastplayerpicked link - it had an clear method for entering/editing values of keepers - but it didnt appear to effect the generated values in any way.

any other suggestions of sites that address this?

[james33]

GETTING BACK TO THE ORIGINAL POST
it seems as though we've not yet come up with a (free) software or website that will calculate auction values including a given set of parameters including keepers (ie, incorporating inflation into the results)? it sounded like we were close with the lastplayerpicked link - it had an clear method for entering/editing values of keepers - but it didnt appear to effect the generated values in any way.

any other suggestions of sites that address this?

I haven't seen a site that addresses this, but I just do this in Excel for my own purposes, and it hasn't proven to be terribly difficult.

Column A = Player Name for the players you think will be either kept or taken at auction
Column B = Dollar Value of each Player (you could use any site's dollar values or your own or an average across several sites)
Column C = The Salary of Any Keeper, or blank for players who are not kept
Column D = Auction value, calculated as follows:
---the column C value for a player if a keeper.
---if not kept and column B value is less than 6, the column B value.
---if not kept and column B value is 6 or more, the column B value multiplied by a single inflation factor

Keep trying different values for your single inflation value until Column D sums to $3,120 (12 teams X $260/team), or whatever the total auction dollars to be spent is.
Sometimes I also do run this separately for pitchers vs. hitters in case inflation in a league is different for one vs. the other.