# Win Your Fantasy Football Auction Draft: Calculate the Optimal Players to Draft with this Shiny App in R

110In this post, I use a Shiny app in R to determine the best possible players to pick in a fantasy football auction draft. The app includes the most accurate fantasy football projections available, and calculates a robust average of more sources of projections than any other website using a robust average (see here for a list of the sources of projections). You can even choose how much weight to give each source. Based on your league settings, it determines which players you should draft to maximize your starting lineup’s projected points. It also allows you to change your risk tolerance to avoid picking risky players. Best of all, the app updates the selections automatically with your inputs, and you can download the data for yourself. So let’s get to it. Here’s a more thorough description:

### How it Works

First, I use a script to scrape player’s projected points from numerous sources using R. Our projections include projections from FantasyPros, an ideal source of projections because it averages across numerous sources of projections, and is the most accurate source of projections. Second, I scrape average auction values from Yahoo, ESPN, and FantasyPros. Third, I calculate players’ risk levels, as defined by the standard deviation (uncertainty) around the players’ ranks and projections across sources. Note that risk is standardized to have a mean of 5 and a standard deviation of 2. Fourth, I take user inputs on league settings to calculate projections that are custom tailored for your league (e.g., based on how many points a passing TD is worth in your league). The projections are based on a weighted average using the source weights input by the user. If the weights are all 1, a robust average is calculated that is less affected by outliers (crazy projections).

Fifth, I take the average auction values (AAV) from the source you select (e.g., Yahoo, ESPN), and normalize the players’ costs relative to the cap available and to the number of teams. In line with an unpublished analysis from Harvard on the optimal draft strategy, I apply a 10% premium to the top players and a 10% discount to the lower-tiered players. According to the study, the majority of one’s excess cap should be spent on the starting lineup, and you should spend less on bench players. This is because bench players do not contribute much to the team’s projected points. Although bench players are important in the case of a starter’s injury or under-performance, it is generally best to draft starters with minimal risk to ensure solid, if not superior, performance. On the other hand, one should accept more risk with bench players because such a move has a low risk, high reward potential. This 10% adjustment results in a **projected cost that reflects the cost you can expect and should be willing to pay for a player if he is selected by the optimizer**. Note: The projected costs don’t always map onto the projected points because the projected cost is derived from the AAV of the site you selected and not the projected points. That’s because your draft costs will be in large part driven by your site’s AAV, which greatly influences what other people in your league will be willing to pay for each player.

Then, based on how many starters you need for each position, your cap available, the players already drafted, and your maximum risk tolerance, I use the Rglpk package to find your optimal starting lineup by selecting the remaining players available that maximize the starting lineup’s sum of projected points while meeting all of the constraints. For a similar execution using Excel’s Solver function, see here. See below for a description of the possible user inputs.

### User Inputs

Players You Drafted: Select all players you’ve already drafted (can type player’s name or scroll to select).

Cap Spent: How much cap you spent on the players you drafted.

Other Players Drafted: Select all players that other teams already drafted (can type player’s name or scroll to select).

Auction Values: Source of average auction values to use in calculating projected cost.

Number of Teams: the number of teams in your fantasy league.

League Cap: the cap available for each team at the beginning of the draft.

Total Number of Players on Roster: the total number of players on roster, including bench players. This is used to determine your available cap (it assumes you must spend at least $1 on each bench player).

Scoring Settings

Passing Yards Per Point: How many passing yards are worth 1 fantasy point?

Points Per Passing TD: How many points is each passing touchdown worth?

Points Per Passing INT: How many points is each interception worth?

Rushing Yards Per Point: How many rushing yards are worth 1 fantasy point?

Points Per Rushing TD: How many points is each rushing touchdown worth?

Points Per Reception: How many points are receptions worth?

Receiving Yards Per Point: How many receiving yards are worth 1 fantasy point?

Points Per Receiving TD: How many points is each receiving touchdown worth?

Points Per Fumble: How many points is each fumble worth?

Maximum Risk Tolerance: Selects the maximum risk allowed for any player to be considered for inclusion in the optimal starting lineup. Players’ risk levels have a mean of 5 and a standard deviation of 2.

Lineup Settings

Enter number of starters needed from each position.

Weights

How much weight to give each source of projections when calculating projected points. If all weights = 1, a robust average is calculated. Otherwise, a weighted average is calculated. For instance, if you want to exclude ESPN projections, you would give them a weight of 0. If you want to give Yahoo projections twice the weight of CBS, you would give Yahoo a weight of 2 and CBS a weight of 1.

### The Shiny App

Below is a preview of the Shiny App for optimizing your draft by picking the players that maximize your starting lineup’s projected points. For the actual Shiny App, go to: http://fantasyfootballanalytics.net:3838/Auction Draft/

Great app! Does this app update projections-wise when FantasyPros updates their projections? If so, this will be very useful when my draft comes around in a month or so.

No worries at all! We usually draft after the last pre-season game to cover unexpected injuries, etc. so towards the end of the pre-season would be great.

Also, one quick question: After filling in my league settings and downloading the data, I noticed there is a column for both “points” and “projectedPts.” What is the utility of the “points” column? I noticed the risk column is next to “points.” Thanks! Great stuff!

Thanks for your interest! Unfortunately, I haven’t been able to find a way to auto-update the data when FantasyPros updates their projections. I just updated the data for you, though, so it should be up to date. I plan to update the data every few weeks. Feel free to let me know a week before your draft, and I’d be happy to update it beforehand.

Thanks for bringing that to my attention. I just deleted the points column, so now there should be no confusion which is the column for projected points (projectedPts). Hope that helps.

This is excellent. But does it take into account how many teams are in a league?

Thanks Oz. The app does not currently adjust for the number of teams in a league. I’d imagine that the number of teams in a league could affect players’ costs (with more teams leading to higher player costs). This could be a pretty easy fix if you know how to adjust for the number of teams. How would you propose adjusting players’ costs depending on the number of teams in one’s league? The FantasyPros average costs are calculated with 10 teams. Thanks for the suggestion!

Great stuff. Probably a stupid question: What do “cost” and “projectedCost” reflect? It seems to me that if either one is supposed to reflect the auction “value” of a player, all the values should add to about 2000 (10 teams x $200 per team). For the settings I tried, both columns added up to well above 2000.

Thanks for your work! I may too be asking for this to be updated before my draft.

Good question, Justin. The variable “cost” reflects the average auction values from fantasypros (http://www.fantasypros.com/nfl/auction-values/qb.php). The variable “projectedCost” reflects a 10% premium for the top players and a 10% discount for the lower-tiered players, as suggested by the Harvard analysis on how much to bid for players (http://harvardsportsanalysis.files.wordpress.com/2012/04/fantasyfootballdraftanalysis1.pdf). These values are also scaled to your league cap. Does that make sense? I’d be happy to re-update the projections prior to your draft. Just let me know. Glad you find it useful!

Isaac, is there any way to add additional point categories? For instance, my main league gives 0.2 points per rush attempt and decent points for big games, i.e. 300 yard passing and 100 yard rushing/receiving.

Hey Kevin: I could add point categories for any category that has a projection. FantasyPros does have projections for rushing attempts, but does not have projections for “big games”. Would you like me to add rushing attempts as a point category?

This is awesome! Is there any chance you can update the projections this Friday or Saturday? Thanks!

Thanks for your interest Ben! I’ll plan to update the projections on Saturday (I’m traveling on Friday). Thanks!

2 Things.

1. To calculate for different league sizes can be done but it is a little harder. Basically all of your values will go up because there is more overall money available but the same amount of players. You would have to build up from the bottom. take all the money that will be spent, subtract out your dollar players and then allocate the rest similar to how it is now. An easier way would be to have a percentage increase for the higher percentage of money available. Probably could acheive the same goal by just increasing the Budget to what it would be for the new number of teams. (240 for a 12 team which is 12*200/10)

2. How are you generating the list of 7 starters on the right side of the calculator? I want to do something similar in an offline excel file.

Nevermind I see how to do it in Excel now. Thanks. But I hope you take my advice on the larger leagues it would help a lot

Hey McConnell and Ben, updated the draft optimizer app with the latest FantasyPros projections and I revised the app to take into account the number of teams in one’s league (http://fantasyfootballanalytics.net/2013/08/update-to-draft-optimizer.html). Thanks!

Thanks again!!!

Isaac, great tool! Thanks so much for sharing. I have one request that would be very helpful for me. We do an auction keeper league, so each of us starts out each draft with up to four players and a certain amount less than the total cap. For instance, we have 10 teams, 20 roster spots, and $250 cap each. But there are 39 players being kept for a total value of $893. So the total cap available to be used on the remaining players should be deducted by that amount to get a better estimation on each player’s worth. Would it be possible for you to account for this in the app? Basically you would need to ask how much total cap money has been spent prior to the draft and then figure the remaining values accordingly. That would be awesome if you could that. Not sure if you would have time to add this beforehand, but my draft is next Saturday. Thanks so much for all the work you have done already!

Well, the available cap will be different for every team, but if I understand what you are saying, I would take the total cap available and divide by 10? So the original cap is $250 per team (or $2500 for the league). The total cap used is $893. So $2500-$893=$1607 / 10 = $160.7. So I would enter $161 in the available cap? Thanks for your time.

Ok, thanks for the clarification. Yeah, that’s not doing what I was hoping it was. When you have some teams you acquired players at less than their going rate (for instance, I picked up Alfred Morris for $1 off waivers last season), then you need to have a way to evaluate the value of the remaining draftable players. The total available cap (all teams) need to be considered in relation to the players still available. Naturally, a premium will be placed on the remaining players. I am just trying to figure out a way to get a proper value on the remaining players. See this for an idea of what I am talking about: http://espn.go.com/fantasy/football/story/_/page/NFLDK2K13_inflation_calculation/how-calculate-inflation-fantasy-football-auction-keeper-league.

Hey Scott. Thanks for your interest! If I understand correctly, what might be easiest would be to set your league cap to your “available cap” when accounting for the keeper players. Then it should give you what you’re looking for. In other words, if your 4 keeper players cost $120 total, input your league cap as $130 for the draft optimizer (250-120 = 130). Does that make sense?

Hey Scott, each team has different numbers and costs of keeper players, right? If so, you’ll have to calculate the cost of *your team’s* keepers and subtract that from 250 to determine your team’s available cap. If each team has the same number and cost of keeper players ($893/10 = 89.3), you’ll subtract 89.3 from 250 to get your team’s available cap (160.7).

This comment has been removed by the author.

Isaac, thanks for this tool, it’s fantastic. If I understand this correctly, can I update while drafting in my actual league? Also, any plans to allow input on who was drafted and for what price to see if that would update the projected cost?

Hey Hank, yes, you absolutely should be able to use the tool and update it while drafting. Just add the players that you draft and that others draft, and the cap that you spend, and it will update with the best available remaining players.

It’s a very intriguing idea you have to update the projected costs based on the costs of other players who are drafted. I imagine the user would need to input every player and what cost they were drafted for, so it would be pretty tedious. I’d have to think about the correction that would be applied to the remaining players. Any suggestions? Nevertheless, the tool will update with the best available players for your remaining cap space.

Thanks for the updates. also picking selected players is much easier now too!

Hi Isaac, this looks like a wonderfully useful tool. I was just about to try and make something like this myself using R.

Right now I get a page that says Not Found at URL http://spark.rstudio.com/dadrivr/FantasyFootballDraftOptimizer/session/8cb41f40d1e224ad3d50293f587a16d4/download/downloadData every time I try to run analytics (the session id changes each time)

Use this website to access the tool:

http://fantasyfootballanalytics.net:3838/Auction%20Draft/

If you want to download the data, press the Download Data button at the bottom of the page.

Just updated the projections for those with drafts coming up. Good luck!

IM DRAFTING NOW AND IT BROKE HALFWAY THOUGH!! F my life….

I get “An error has occurred / The application exited unexpectedly.” And then a bunch of messages that “data length exceeds size of matrix”

I had the same thing happen to me as Anonymous

Yup something is borked with it now.

Warning in matrix(align.tmp[(2 - pos):(ncol(x) + 1)], nrow = nrow(x), ncol = ncol(x) + :

data length exceeds size of matrix

So sorry, man. What’s the problem? It’s working for me… Hope your draft ends up okay.

Thanks, and sorry for the error. I think I pinpointed the problem and fixed it. Let me know if you run into any other errors. Thanks!

Draft is tonight and it seems that its not working. or am I the only one? It was working perfectly fine an hour ago. and its an awesome tool! Thanks Isaac!

Hmm, it might be a server issue. I’ll try to contact the server admin. Hope it stays up and running for your draft tonight. Good luck!

Isaac… You are awesome! Thank you for this great tool. Have you considered adding a column in the spreadsheet for bye week? That would be helpful to have during a draft.

I just received this response from the server admin: “We recently switched the way we communicate between Shiny Server and Shiny, to use Unix domain sockets instead of the normal way (TCP connections). When testing under load we’ve found Unix domain sockets to be less stable after all, so we’ve gone back to TCP connections–this should be deployed to glimmer and spark on Friday of this week. It should fix this problem.”

Bottom line: it should (hopefully) be fixed Friday.

Thanks man! I haven’t included bye weeks for two reasons: 1) FantasyPros does not include bye weeks on their projection pages, so I would have to merge them from somewhere else, which is more complicated and could be messy, and 2) it makes sense to draft the best possible team regardless of bye weeks because you don’t know how the bye weeks will help or hurt your chances. For instance, it might actually be beneficial for all of your starters to have the same bye week. Yes, you might lose that particular week, but you’ll be more likely to win all of the other games. Nevertheless, I might consider adding them if I can be convinced they’re useful for picking the best team.

Can you share the formula you are using for projected cost? I’d like to make on the fly adjustments based on how draft money has already been spent. Awesome tool Thanks!

It applies a 10% premium to the highest-tier players and a 10% discount to the lowest-tier players. It adjusts the projections according to the number of teams and the league cap (from the original projections derived from a league with 10 teams and a cap of $200).

For those looking for an optimizer tool for snake drafts, I just finished creating one.

Shiny App: http://fantasyfootballanalytics.net:3838/Snake%20Draft/

Accompanying blog post: http://fantasyfootballanalytics.net/tools/fantasy-football-snake-draft-optimizer

Awesome tool Isaac! Any way you can update the rankings today? I have a draft tonight.

Finally draft day and its down

Not sure what the problem was.. I might need to look into a new server for hosting. Any suggestions? I’d prefer a free one.

Hostgator.com is pretty good

We updated the app and placed it on our servers. It should be more reliable now.

This tool has got me into a championship game this year! 1 game to go!!! Amazing job isaac!

Congrats Homy! How’d you do in your league championship? We’ll be updating the data for the app this offseason to prepare for next year’s draft. Good luck!

Isaac,

How easy would it be to update with the 2014 Preseason Rankings? I have a Draft tomorrow I’d like to use this with. Additionally, can this be customized with a 32 team league?

Hey RGH,

I’m planning to update the data when sites make their 2014 projections available (still waiting on ESPN, NFL.com, and FantasyPros). The app uses projections as opposed to rankings (you can derive rankings from projections but not other way around), so unfortunately I wouldn’t be able to use preseason rankings for the app. Check back soon, though, for update data based on preseason projections. Regarding a 32-team league, the app allows you to specify how many teams, so it feasibly could be customized, but I don’t know how well the cost values would extrapolate to a 32-team league. Might take a look at it to seem if the values seem reasonable.

-Isaac

[…] and many users have contributed packages for performing specialized functions. For example, this fantasy football draft optimizer uses the Rglpk package to find your optimal starting lineup of players that maximizes the […]

[…] Win Your Auction Draft […]

[…] Win Your Auction Draft […]

Isaac,

Exactly the tool I was looking for (you should try to monetize it). What you may consider factoring in is Yahoo’s and ESPN’s projected auction values which is what many auctioneers will be referencing when drafting on one of the two fantasy hosting services. Then you apply the inflated values for the top end players. Again, awesome tool and I will have this up and running during my draft this year!

Great point, Ryan. I’ll try to find a way to incorporate ESPN and Yahoo projected auction values. Glad you find it helpful!

Hey Ryan, just added Yahoo, ESPN, and FantasyPros auction values to the apps. Cheers!

-Isaac

Awesome Isaac. Can’t tell you how impressed I am. Not sure what your opinion on this is or how it would be factored into your calculations but check out ESPN’s 2013 analysis on fantasy consistency (http://espn.go.com/fantasy/football/story/_/page/consistency131231/russell-wilson-wes-welker-prove-consistent-again). I think consistency is extremely overlooked in fantasy football when you’re competing in head-to-head leagues. Cheers

*week-to-week consistency

Week-to-week consistency is an interesting topic, and it was the topic of my recent simulation:

http://fantasyfootballanalytics.net/2014/07/weekly-variability-simulation.html

Not sure about the best way to incorporate it into the app because a simulation takes time to run, and its findings would depend on the inputs (e.g., your league settings).

1 quick update – please allow for 0.5 PPR

Are you sure it’s not allowing you to enter 0.5 PPR? It works for me (even though the cell is highlighted red, the calculations appear to update correctly).

The 0.5 does work – my apologies. However, now the ‘Players You Drafted’ and ‘Other Players Drafted’ are not working properly. When I click to type in there is only 1,2,3 etc. and not the list of names as before then if I type in a name it gets deleted after entered. I tried using the optimizer tool on both Chrome and Safari with the same result.

Another quick comment – could you order the player output by projected cost rather than first name?

Just fixed the “Players Drafted” bug to show player names instead of numbers. Let me know if you still have problems with it. Is there a reason for sorting by cost? I thought it’d be best to be able to see all players sorted by name so you can easily find any player. If you want to find players the fastest, typing in their name is probably best (it will autofill based on what you type). What are your thoughts?

My thought was that auction drafts typically start with the highest projected players and you are thinking in a top down approach. If I dont win this player, who is the next best player I can spend my money on?

Do you mean sorting the player list or the table of players in the output? If the former, that’s easy. If the latter, I found a way to do that, but I won’t be able to release the update until we upgrade our Linux server (it requires a newer version of R packages). We’re looking to pay someone to upgrade our Linux server. Hopefully this happens soon, but if you know someone qualified let me know!

Just updated the app with sortable projections in the output. Is that what you’re looking for?

I’m set to release a fairly important update to the apps, but I’m having some linux trouble. If you’re good with linux, please contact me.

Just updated the app with sortable projections. Cheers!

saw that. thanks!

I see that Defenses are not taken into account. I have won weeks based on a strong defense pulling through for me. Would this not be a valid addition to the analytics?

Hey Ricky,

The reason why I haven’t included Defenses is because 1) very few sites have projections (not just rankings) of defenses, 2) it’d be a lot of scripting (i.e., time) to include them here and 3) one should generally draft kickers and defenses last because they are the least predictable (http://fantasyfootballanalytics.net/2013/07/expected-points-by-position-rank-in-fantasy-football.html). That said, if someone is willing to help me include them, I think it’d be a great addition.

-Isaac

Thanks!

is there any way in the provided data to incorporate consistence? maybe use the stand dev or risk column somehow just a gauge/ reference point?

Hey Brent, the standard deviation reflects the variability/consistency from source to source in a player’s projections. If you want an estimate of a player’s week-to-week consistency, we conducted a simulation to calculate that (http://fantasyfootballanalytics.net/2014/07/weekly-variability-simulation.html). It would be difficult to incorporate this into the app because a simulation takes time to run, and its results would depend on your league scoring settings (so it wouldn’t be very user friendly because the app would be slow). That being said, the code is on GitHub, so you can run it for your custom league settings to get an idea of players’ week-to-week consistency for your league settings.

This is fantastic! Great work!

Im a little confused on the snake draft “typical replacement” options. Can you clarify exactly what those are please?

Also, wouldn’t it matter how many starters/teams there are in the snake draft in order to calculate best picks?

Hey Ricky, the “typical replacement” is for snake drafts (the app on this page is for auction drafts). Here’s info on the snake draft app:

http://fantasyfootballanalytics.net/2013/09/win-your-fantasy-football-snake-draft.html

Basically, “typical replacement” is the position rank of each position that you want to normalize each player’s score against. For more info on a player’s value over a typical replacement player (VOR), see here:

http://fantasyfootballanalytics.net/2013/04/win-your-snake-draft-calculating-value.html

The snake app shows the two best players remaining at each position. You can then pick which player to draft based on which players have highest VOR and which positions you need. Not sure why the best remaining players at each position would depend on roster/league size.

Hope that helps!

[…] Win Your Auction Draft […]

Hi Isaac,

Thanks for all of the work you’ve put in to this and for not putting it all behind a paywall. Your effort is much appreciated.

My question relates to the AAV. Why are there multiple options and why does it change the results so much? Shouldn’t the dollar value for each player be calculated by league settings, dollars remaining, and available players?

Hey Justin,

Good question. The multiple options are to select which auction values you want to use. If you’re drafting in Yahoo, I’d use one of theirs. If ESPN, use theirs. The short answer is that each site provides considerably different auction values, so I’d recommend using the site you’re using for your draft. Yahoo provides 2 different ones (AAV and projected value). I like to use the average of the two in my personal drafts, but that’s personal preference. You’ll want to pick whichever will most closely align with the costs people will bid in your draft.

The auction values are further scaled to your league cap and the number of teams in your league. The top players receive a 10% premium and the worst players get a 10% discount.

Hope that helps!

-Isaac

Doesn’t using the dollar values from just one site nullify the aggregation of the different predictions? Shouldn’t the value you’re willing to pay for a player be totally dependent on his projected stats?

What *you* bid should absolutely depend on a player’s value. There are two problems in this context: 1) it’s difficult to determine a player’s value in terms of cost because it’s not a linear transformation. Here’s my attempt to calculate a bid-up-to value: http://fantasyfootballanalytics.net/2013/08/calculating-bid-up-to-values.html. 2) More importantly, what *others* will bid will depend largely on the site you’re using. As a result, you wouldn’t want the draft optimizer to tell you to draft a player at a cost lower than you will be able to be draft him for.

I’m still confused. So, you use projections to get projected fantasy points. Then you use the auction values as the cost of the player for the optimizer? Stately differently, the projections determine their points and the auction values represent the cost to get that player. You’re not generating an independent auction value based on projections before generating the final auction values.

If that’s true, then the site’s auction values serve as the base, correct? None of your values will be lower than ESPN’s auction values?

You are correct that I am generating the projections and costs separately. The costs come from the different sites (Yahoo, ESPN, etc.). I then scale them to the number of teams in your league (more teams = higher cost) and your league cap (higher cap = higher cost). Then, as suggested by a Harvard study (http://harvardsportsanalysis.files.wordpress.com/2012/04/fantasyfootballdraftanalysis1.pdf), I apply a 10% premium to top players and a 10% discount to bottom players. The costs could be lower than ESPN’s if you have fewer than 10 teams, a lower cap than 200, or are looking at bottom players. Hope that clarifies!

If others are “overbidding” based on the stats a player is projected to generate, how could that be an optimal selection? Regardless of the site you’re drafting on, every player has to have a projected value based on your leagues scoring rules, roster size, and budget.

Yes, players have a value. You can compute a cost for this value. If you rely on bidding on players at their value-based cost, however, you won’t get the best players necessary to maximize your projected points. Don’t take my word for it. See here for more info: http://harvardsportsanalysis.files.wordpress.com/2012/04/fantasyfootballdraftanalysis1.pdf. To help you gauge a player’s true value, I recommend using the “bid-up-to” value.

Upgrading our server to handle more traffic. Apps will be down for a bit. Will get the up as soon as possible.

-Isaac

Apps are back up now!

When you mention “Bid up to” earlier in your comments for deciding how much to bid on a player, you are referring to the projectedCost on your stats? We shouldn’t pay more than the projectedCost (generally speaking), but getting someone for less than that variable would be a “success”?

thanks- great work

Hey JM,

The Bid-Up-To value is calculated by one of my scripts:

http://fantasyfootballanalytics.net/2013/08/calculating-bid-up-to-values.html

It calculates the optimal value for each player. It is different than the projected cost, which is the cost you can expect each player to go for in your league (i.e., what you would have to bid in order to win a player). Unfortunately, there’s not an easy way to add Bid-Up-To value to the app because it is a simulation that takes several minutes to run, so it would be too slow for users in the app. I don’t think the Bid-Up-To value is crucial, however, because if you know about how much each player will go for, you can estimate fairly well the best lineup to draft given your league settings and constraints.

Hope that clarifies!

-Isaac

Excellent tool for fantasy. One question I have is the projected cost vs. position rank. It appears that in some instances lower ranked players have a higher projected cost. If the definition of projected cost is “cost you should allot to pay for a player” why would one pay more for a lower ranked player?

For example, Cam Newton is the 8th ranked QB with a projected cost of $11, while Andrew Luck is the 4th ranked QB with a projected cost of $7. Why would one allot more money to pay for Newton when he is projected to do worse than Luck?

Hey Mike,

Great question. The projected cost is how much you can expect players to go for in your league. It comes from your site’s AAV. As a result, it is not necessarily how much you should bid for a player (that’s the Bid-Up-To value, see here: http://fantasyfootballanalytics.net/2013/08/calculating-bid-up-to-values.html). Obviously, for your league, Luck is a better value than Newton. The app is just telling you that Newton is likely to cost more than Luck in your league’s draft according to the site’s AAV that you selected. As a result, the tool would be more likely to select Luck than Newton because he contributes more points to your lineup with a lower cost.

Does that make sense?

-Isaac

These tools are simply fantastic. One question I have is whether there is a way to figure out bid-up-to value using the projections or auction optimizer without using the optimizer loop in R? I have absolutely no experience or understanding with R, so I was hoping I could use the other tools available to figure out a rough estimate for bid-up-to value.

Any ideas on that front? Again, thank you again for putting these tools out there and making them freely available. Easily the best analytical tools I’ve seen.

Hey Doug,

Many people have requested this feature, so I’ve spent some time thinking about how to incorporate it. One way would be to calculate the Bid-Up-To value only for the selected players (and not all available players). That would be a much faster loop than estimating it for all players. I’ll look into this when I get back from travels.

Thanks for the suggestion,

Isaac

It appears that when I download the results I get something very different from the top 5 preview shown on the page. Would you expect QBs to take all the top spots? Did I do something strange in the settings to make that happen?

Also, if you don’t have a cost cap in your league, is there a way to deactivate that?

Alright – I get that the downloaded sheet contains everyone, not the suggested drafts, but I’m still confused that it appears that all the top values/costs/ranks go to QBs – how do I need to sort the sheet to make sense of it? And what about teams where there is no cost cap – just a draft order consideration?

Hey Josh,

If your league doesn’t have a cap, then you probably want the Snake Draft optimizer (not the Auction Draft optimizer):

http://fantasyfootballanalytics.net:3838/Snake%20Draft/

If you’re just looking for rankings or projections, see here:

http://fantasyfootballanalytics.net:3838/Projections/

The rankings depend on your league scoring settings and your “typical replacement” selections. If your league scoring settings weight QBs highly, then the rankings will favor QBs. The default scoring settings don’t appear to favor QBs, though (Peyton is the only QB in the top 20). This suggests to me that some of your inputs (e.g., league scoring settings like points per passing TD) are giving more weight to QBs.

Hope that helps,

Isaac

That’s really helpful – thanks. It does seem partially to do with the league scoring rules, but also that the page I was looking at doesn’t seem to have values for some of the things we add points for (field goals, defensive stuff that isn’t safeties)… I’ll look more closely to see if I’m missing anything.

Isaac, I have my auction draft on Sunday – when will you be updating the player projections and AAVs? Thanks again!

Ryan

PS. Can I buy you a beer via Venmo?

Hey Ryan,

Hope to update the data Friday. Thanks for the offer for the beer! How about I make you a deal.. Tell some friends about the site (obviously friends who aren’t in your league), and we’ll call it even.

How does that sound?

-Isaac

You got it! Thanks and if it wouldnt be too much of a bother can you comment here if you update them friday- will give me a little more peace of mind Sunday when I’m relying on this app to win my league

Just updated the data and added return TDs to the draft apps. Go win your draft!

Cheers!

-Isaac

Thanks for all your work. I’m a huge fan! I’ve got one suggestion, though:

The app doesn’t seem to factor in the roster size in calculating auction values — instead, it seems to allocate 99% of the cap to a starting lineup, in which case a team would have zero dollars for bench players. Is that the intention or am I doing something wrong?

We have a larger roster of 15 players in our league, and spending 5-15% of the cap on bench players has been a successful strategy to mitigate injury disasters, stash breakout candidates, and make weekly substitutions based on matchups.

Sorry, I just looked and we now have a roster of 20 players, and the minimum cost for the additional 13 players would be 5% of the cap, and 10% allocated would probably enable an owner to get the best pickings of players who would other be waiver wire material.

Hey Justin,

The default settings leave $1 for each bench player, and find the starting lineup with the highest sum of projected points with your remaining dollars. That’s because bench players don’t contribute to your starting lineup unless they outperform starters. If you want to allocate more dollars to bench players, put the extra amount you want to reserve in the “cap spent” section, and it will be taken out of the pool you can spend on starters. The projected costs do take into account roster size. They are scaled proportionately. It’s not clear to me how accurate this scaling is with super large leagues (e.g., 20 players), however. Take a look and see if the costs seem reasonable. If not, you may want to enter fewer teams in the app until the costs seem appropriate.

Hope that helps!

-Isaac

[…] Win Your Auction Draft […]

Great website! I included all the players already drafted, but is there anyway to factor in the remaining cap space of all other teams? Each teams keepers carry different values.

Hey Jon,

There’s not currently a way to do this in the app because the auction costs are derived from average auction values (AAV) and not other teams’ remaining cap per se. I can consider adding this, though I suspect it would be fairly complicated.

-Isaac

Hi Isaac

I can’t tell you how much I love this website. I have played with the auction app everyday for the last two weeks in preparation for my draft this Sunday. I too am a statistics nerd, and I love the fundamentals you’ve applied to fantasy football.

I have a couple of questions for you:

1. Will you be doing any projections updates between now and Sunday? I want to make sure I have all the latest information on all of my printouts.

2. It looks like the apps are down right now. You can promise me that they will be working at 1 PM Central on Sunday right?

Hey Peter,

1) I’ll be traveling this weekend, but I just updated them last night, so they should be ready to go.

2) The apps are back up now. Sorry about the downtime! That’s one of the downsides of running a free operation — we don’t have the largest server capacity. We are looking into creating standalone software so people don’t have to rely on an internet connection or on our server. Should be good for Sunday, though.

Good luck in your draft!

-Isaac