Win Your Fantasy Football Draft with These OpenCPU Apps: 2014 Update
111By popular demand, I updated the fantasy football draft optimizers with 2014 projections. There are two draft optimizers (one for auction drafts and one for snake drafts). The optimizers identify the best players to draft (highest projected points within your risk tolerance) given your league settings. The optimizers include the most accurate fantasy football projections available. Another app calculates custom rankings and projections for your league. Our projections combine more sources than other website using a robust average. You can even choose how much weight to give each source. There is also an app to identify sleepers (players who are likely to exceed their statistical expectations and have a breakout season).
Here are the OpenCPU apps:
- Auction Draft Optimizer
- Snake Draft Optimizer
- Identify Sleepers
- Projections/Rankings for Your League
Hey Isaac, I love your site. These links to the shiny apps aren’t working though. I can only access cached versions from 6/13/2014
Hey Lewis,
We had a high level of traffic and had to reset the server. Could you try the links again? They’re working for me now. Let me know if they’re still not working for you. In the meantime, we’re looking into options to increase our server capacity. Let me know if anyone has suggestions.
Thanks,
Isaac
Hey Isaac, super excited to have found your site. All this data looks great. Any word on when your Auction Draft optimizer will be updated with 2015 projections? It’s the best tool yet I’ve found!
Thanks Greg! We’re hoping to release it in about 3 weeks. Stay tuned!
-Isaac
Will the snake projections be updated as well?
Hi Michael,
You can use this for your Snake Drafts:
http://apps.fantasyfootballanalytics.net/projections
-Isaac
In response to a user’s request, I moved the “Players Drafted” section of the draft optimizer apps to the top of the screen to reduce the need for scrolling during your draft.
Cheers!
-Isaac
Still not working for me, thanks for working on this
Sorry for the hassle, Lewis. I’m not sure what the problem is. Could you try it on a different computer or browser? It’s hard to troubleshoot if I can’t reproduce the problem.
Also, you might try clearing your browser cache.
Ah yes, I think clearing the cache did the trick. Sorry, should have tried that right off. Thanks.
One feature that would be great would be the option to include only the desired sources of projections (and/or weight the sources as you mention in one of the articles on your site). I would actually like to do this myself by simply modifying your code on github, but I see that some of the site projections are password protected. Great work, thanks!
Great suggestion, Lewis. I just updated the Projections App to calculate weighted averages of projections by user-defined weights:
http://apps.fantasyfootballanalytics.net/projections
Should I do this for the other apps, as well?
Excellent! Thanks. I think it would be useful for the draft optimizers as well, certainly.
Just updated all of the apps to allow weighted averages. Hope that helps!
Great stuff as always! Been using your app for a year. One question, though. Your projections spit out some odd rankings and projections for me. The top 17 rated players in terms of projected points are all QB’s (with Ben Roethlisberger at #17 overall – crazy!). Here’s two screenshots of the settings I input in your auction draft optimizer:
1) http://i.imgur.com/zwk1qyM.png?1
2) http://i.imgur.com/5V5gSWL.png?1
Last year, your projections had the highest rated QB as 8th overall in projected points (Brees). The only league settings that have been changed in my league is the addition of one more starter position (RB/WR/TE) while keeping the roster at 15 (so we lose a bench guy). I’m wondering, what’s going on here? It just doesn’t make sense, especially in a league where QB TD’s are worth only 4 points and there’s only 1 QB starter.
Thanks!
Crap – I just saw this link: http://fantasyfootballanalytics.net:3838/Projections/
That gives me MUCH different results (highest rated QB is #10 – Manning). Crap!
I was using this link from last year to calculate my initial projections: http://fantasyfootballanalytics.net:3838/Auction%20Draft/
Hey James,
It looks like you found the link to our projections (http://apps.fantasyfootballanalytics.net/projections), which calculates a robust average across numerous sources. Those projections are different than the ones in the Auction/Snake Draft Optimizers (which use FantasyPros projections).
Hope that helps!
-Isaac
FYI, the Auction/Snake Draft Optimizers now calculate a robust average across numerous sources just like the projections app. You should get the same projections with both apps now.
Absoultely love the apps.
My favorite part is the inclusion of an upside metric, as it’s imperative to shift from expected projection when drafting starters, to potential upside when drafting your bench.
Quick question, do the apps automatically pull in the latest released projections from the sites? I only ask because the projections haven’t changed the past two weeks or so. Clearly with developments like David Wilson retiring, etc., I would expect projections might be changing in the coming weeks.
Best Regards
Hey Elliot,
Thanks–I appreciate the feedback. I will update the projections soon (tonight or tomorrow). Will send out a notification on Twitter when they’re updated (https://twitter.com/FFAnalyticsNet).
I’ll also try to update them fairly regularly now as draft season approaches.
-Isaac
Just updated the projections. They are only as up-to-date as the sources (ESPN, Yahoo, etc.) are. Not sure if all sources have released updates since David Wilson’s retirement.
Is there a way to add the ability to have position specific scoring for the projections? For instance, we use different scoring for TE than we do for other positions (1pt. per 5 yards instead of 1 pt. for 10 yards for other positions). Any ideas?
Thanks and I’m really digging the site.
Hey NS, interesting question. That would be do-able, but the only way I can think of implementing that would be to have separate dialog boxes for every combination of stats category (10) and position (4): 10 x 4 = 40 different dialog boxes. That seems excessive. Any ideas?
The leagues I’ve seen only really change the value for yards. Rec. and rush yards x 4 positions – the two already there would net 6 more boxes. If TDs are added it would be another 6… Better than 40, but I don’t know how you feel about it.
If it is an aesthetics thing, you could just keep it as is for the default and add a selector box that would make the additional scoring boxes visible when needed… Just a thought.
Keep up the good work.
That sounds reasonable. I’m in my dissertation year, so it’s unlikely I’ll be able to get this before this draft season, but I’ll definitely put it on the to-do list.
Thanks. I appreciate it and good luck.
I was curious why there is no list of how many starting QB’s etc. on the snake draft optimizer and there is on the auction draft optimizer. Also I was wondering why on both there is already a listed team not one which I can add.
In the auction draft app, the number of starting players at each position is used for calculating the best starting lineup to draft (most projected points within cost and risk tolerance).
In the snake draft app, the top two players at each position (in terms of ‘value over replacement’) are presented. This allows you to pick which position to draft based on which positions show the fastest decline in VOR. Unlike in auction drafts, where every team is able to bid on any player, in snake drafts you can’t necessarily know which players will be available during your next pick, so we can’t calculate the ‘optimal lineup’ without knowing who else is available (i.e., you should pick players with the highest value relative to the remaining players). For more info, see here: http://fantasyfootballanalytics.net/2013/09/win-your-fantasy-football-snake-draft.html.
Not sure what you mean by there already being a team listed that you can’t add. I updated the combo boxes so you can type or select multiple players who’ve been drafted to remove them from the list of remaining players. Hope that helps!
I just updated the apps to allow the user to pick the source of auction values (Yahoo, ESPN, or FantasyPros). Hope that helps!
I was using the Auction Optimizer today, and it looks like something happened to the “Players You Drafted” and “Other Players Drafted” fields. Yesterday you could populate those fields with player names. Today it is only populated with numbers 1-780. I’m not sure if this is some ID number.
Hey Peter, just fixed the “Players Drafted” bug to show player names instead of numbers. Let me know if you still have problems with it. Thanks!
I am still seeing numbers instead of actual player names. I am using Chrome by the way.
Just fixed this for the Snake draft app. Let me know if you’re still having problems with it. Thanks!
Hi Isaac, great stuff. I’ve looked through your R code for the League Setting.R script and see how I can modify the scoring system for our particular league in that file which is sourced in the Draft Day.R script. My question is how would the draft optimizer take into account scoring rules for Kicking and Defense/Special Teams? For instance, our league has the following points: Thanks for your help!
#Kicking:
#PAT Made: 1 point
#FG Made 0-19: 3 points
#FG Made 20-29: 3 points
#FG Made 30-39: 3 points
#FG Made 40-49: 3 points
#FG Made 50+: 5 points
#Defense / Special Teams:
#Sacks: 1 point
#Interceptions: 2 points
#Fumbles Recovered: 2 points
#Safeties: 2 points
#Touchdowns: 6 points
#Kickoff and Punt Return Touchdowns: 6 points
#Points Allowed 0: 10 points
#Points Allowed 1-6: 7 points
#Points Allowed 7-13: 4 points
#Points Allowed 14-20: 1 point
#Points Allowed 21-27: 0 points
#Points Allowed 28-34: -1 point
#Points Allowed 35+: -4 points
The draft optimizers don’t include kickers and defense because you should draft them last, as they are the least predictable:
http://fantasyfootballanalytics.net/2013/07/expected-points-by-position-rank-in-fantasy-football.html
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 apps with sortable projections and you can now compare specific players in the projections app. Cheers!
Isaac, love the apps. I’m doing a snake draft and in the eight round when I try and select some players I keep getting this error: duplicate ‘row.names’ are not allowed.
Hey Mike, I think I solved the problem but not entirely sure because I wasn’t able to reproduce it. Let me know if you’re still having problems, and if so, which players you picked that gave you the error. Thanks!
This isnt working for me, I’ve tried to look on my s5 using chrome and a mac using firefox. Do you have any suggestions? I even cleared the cache.
Hey Joe, sorry we had a huge traffic spike, and the apps and website went down. They should be working now (let me know if you’re still not able to access them). We’re looking into ways to prevent this from happening again. Sorry for the trouble!
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!
My buddy is a crazy person and wants to do a 6 team league with 2 QBs AND a QB/Flex. How can I adjust the projections app to help guide where I should take QBs if every team will draft at least 3 QBs?
Use the auction or snake draft app (depending on which type of draft your league uses). Auction draft app would tell you which QBs to draft (you can specify that you need to draft 2 or 3 starting QBs). Snake draft app would tell you who the best remaining players are based on their value. It’s likely that you’ll get best value for QBs later in the draft. The snake draft app would show you this.
Hi, Isaac! Question about the sources.
FantasyPros is a 5-source average of the CBS average, Dave Richard, Jamey Eisenberg, Yahoo, and ESPN. So if we include FantasyPros as one source and then Richard, Eisenberg, Yahoo, and ESPN each as individual sources, are’t we actually giving them extra weight?
(Also, considering CBS is an average of Richard and Eisenberg’s projections, does that mean we’re triple counting Richard and Eisenberg? To eliminate this, maybe should users just remove FantasyPros and CBS as projection sources?)
Hey Scott,
Yes, they would be double counted, but only in the context of FP’s average (so they wouldn’t be fully double weighted — ie., ESPN only composes 1/5th the FP average). You could remove the double counting by giving FantasyPros a weight of zero. The reason I keep the default weights at 1 is to calculate a robust average (not a simple mean) that is not affected by outliers. Plus, FantasyPros had the most accurate projections the last two years (http://fantasyfootballanalytics.net/2014/06/best-fantasy-football-projections-2014.html), and their average is probably toward the middle of the distribution. If you change the weights, a weighted average is calculated as opposed to a robust average. I prefer the robust average for my purposes (it gives less weight to the crazy projections), but you have the option to do however you wish!
Hope that helps!
-Isaac
Interesting. You’re 10x the mathematician I am, so let me ask a couple questions out loud and you tell me why they’re dumb. First, is there a point of diminishing returns in calculating a robust average? As in, why not count FantasyPros twice? Why not count it them three times? Second, if we’re trying to minimize outliers, would it be more useful to simply eliminate the highest and lowest projections for each player?
Now back to actual life 🙂
Good questions. You can absolutely count FantasyPros twice or three times if you’d like! If by “why not count FantasyPros twice” you are asking why I don’t for my own purposes, it’s because (as you and others have pointed out) I’d rather not give too much weight to any individual source unless they are reliably more accurate over time. FantasyPros does appear to be more accurate (because it’s an average of multiple sources), but they are an average, not an individual source (i.e., we’re already double counting the sources underlying their projection). If by “why not count FantasyPros twice” you are asking whether it’s a good idea to, that’s a good question. I don’t know the answer, but it’s a testable question, and we should be able to answer the question next year when we compare the accuracy of all of the sources. In the meantime, you can choose whichever approach suits your goals.
I don’t like discarding the top and bottom projections for many reasons. My key reason is that they might not actually be outliers. Alternatively, more than 1 source on the upper end (or bottom end) could be outliers. Also, what if we only have projections from 1 or 2 sources for a player? There are many reasons why a robust average is preferable to blind rejection of top and bottom values.
Isaac,
I love the site. I am going to see how combining your projections/rankings specific to each of my leagues with the consensus ranks over at Fantasy Pros works out. Doing so in mock drafts has so far yielded teams I’d be pleased with, so I am optimistic/excited for my real drafts.
Anywho, I noticed today that in addition to Terrance Williams. there is also a Terrence Williams listed as a WR for Dallas (note the spelling difference). Perhaps one of your sources is the root of the error?
Also, what are your thoughts on Fantasy Football Toolbox? They claim to have reliable projections because “Toolbox won the 2012 Most Accurate Fantasy Football Rankings Contest awarded by Fantasy Pros.” I like how customizable their league settings are, but I find it odd that for my league with standard scoring other than 5-pt. passing TDs, Calvin Johnson is projected as the #1 overall player followed by Demaryius Thomas and Dez Bryant (and then LeSean McCoy).
Staff Rankings for FF Toolbox finished 52nd out of 132. So a little better than average.
Hey Tim,
Thanks for pointing out the misspelling. Turns out it was WalterFootball’s misspelling. I try to catch them, but I miss some, so thanks for finding it! I fixed the data in the app, so it should be fine now.
Not too familiar with FF toolbox. Unless I’m looking in the wrong places, they appear to only have rankings (not projections). I tend not to focus on rankings because a difference of 10 in rank is meaningless if the difference is 2 projected points. That’s why I prefer projections. You can always calculate rankings from projections, but you can’t go the other way around. As I’ve shown, player rankings are specific to your league settings, so I don’t find general rankings that helpful.
-Isaac
Issac,
I love the site and you have some great stuff here. I was wondering if there would be a way to upload our own values to an Auction based draft? For example, in a 2-QB auction league, the values of the QBs are significantly undervalued when compared to the actual results of a 2-QB auction draft. Or, would it be possible to take 2-QB auction league values from another source?
Thanks and keep up the excellent work!
Hey John,
It’s not currently possible to upload your own AAV values, but this is something I can look into for next season. What other source would you like me to add? Including a link to the source would be helpful. It may or may not be helpful depending on your scoring settings for QBs, but in most leagues, it makes most sense to draft QBs last because you can get good value for QBs later on in the draft:
http://fantasyfootballanalytics.net/2013/07/expected-points-by-position-rank-in-fantasy-football.html
In other words, you might not have to bid that much for QBs in your league if you wait until later to draft them. But if QBs are weighted heavily in your league, this obviously isn’t a good option.
-Isaac
Is there anyway we could get an update on Friday?,Hopefully, most sites will have updated by then after the preseason action. This is a big draft weekend!
Hey Andy,
I’m traveling but I’ll do my best to update it by Friday. Should be able to update it Saturday for sure.
-Isaac
Isaac, do the apps scrape current projections or do you have to update them manually?
The apps take the latest projections available on GitHub. The projections on GH are updated whenever I run the R scripts. Bottom line: I can update the projections in the apps by running the R scripts. The apps don’t scrape the latest projections themselves because that would take too long for the user.
Ooh, okay. Then put me in the “I hope Isaac has enough time to update the GitHub projections Friday night” club 🙂
Thanks, Isaac! We draft Saturday morning, but I don’t want to burden you. Thanks for this tool!
Hi Issac. I’m interested in trying your Snake Draft App, but I cannot get it to work in either IE or Chrome. I’ve deleted the cache per the advice of other users, but no luck. Any help would be greatly appreciated. Thank you!
Hey Jason,
Sorry for the trouble. What’s the error/problem you see when you try to load the app? Any info to help me understand or reproduce the problem would be helpful. You might have to disable ad-blockers (not sure though). Just to make sure you’re using the right link, here’s the link to the Snake Draft app:
http://apps.fantasyfootballanalytics.net/projections
-Isaac
Quick bug report: Changing the PPR or Receiving Yards Per Point values seems to make the snake optimizer app drop a bunch of the top players; namely Jamaal Charles, Jimmy Graham, and all of the top 5 WRs in the Excel sheet.
Whoops, never mind. They were getting cut off by the Risk modifier. That makes sense.
Just updated the data and added return TDs to the draft apps. Go win your draft!
Cheers!
-Isaac
Thanks, Isaac! You’re the best!
Thanks a ton, Isaac! Is this the official weekend update? As in, is it safe to start printing up cheat sheets for Saturday/Sunday drafts?
Yes, go ahead. I might update them this weekend, too, but I suspect not much will have changed (if anything).
I have a question on the auction draft optimizer. How do you calculate the projected cost? Do you take the input auction value from the input site (i.e. ESPN) and adjust for my particular leagues scoring settings? Or does this app factor in value over replacement, risk level, etc…? Thanks.
Hey Ryan,
I added a more thorough description of how projected costs are calculated here:
http://fantasyfootballanalytics.net/2013/06/win-your-fantasy-football-auction-draft.html
Basically, it takes AAV from the site you select, adjusts it to fit your league cap and number of teams, and applies a 10% premium to top players and 10% discount to bottom players. This 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. The projected cost is derived from the AAV of the site you select and not the projected points (or risk, vor, etc.). 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. Does that make sense?
Hope that helps!
-Isaac
Yes, makes sense. Thanks, Isaac.
Doing my first fantasy draft tonight. Looking forward to using this tool. Thank you for making my life so much better.
The link to auction tool is broken.
Here’s the link:
http://apps.fantasyfootballanalytics.net/lineupoptimizer
What’s the problem/error you receive? You might try deleting your browser’s cache.
Is the Snake Draft App down?
Apps should be back up now. Sorry for the downtime!
I was wondering about the possibility of adding more projections sources to your projections app? Some that have been rather accurate over the last few years and aren’t in your app are:
EDS Football / Kevin Hanson (http://www.eatdrinkandsleepfootball.com/fantasy/projections/)
4for4 / John Paulsen (http://www.4for4.com/fantasy-football/rankings/QB/STANDARD)
Traveling this weekend, but can look into this in the future. The EDS Football looks possible. Unfortunately the 4for4 would have to be publicly available in order to scrape and update automatically.
Just added EDS Football projections to our apps.
Cheers!
-Isaac
Cause for concern here: Adjusting only passing TD points (5) and replacement values (QB: 15, RB: 34, WR: 40, TE: 10) results in DeAndre Hopkins and Kelvin Benjamin with awfully high vor values:
47 DeAndre Hopkins WR HOU 34.56 149.04 178.25 1.0628 5.5647 1 120.9 73.9
48 Pierre Garcon WR WAS 33.72 148.2 186.14 2.0316 5.2832 1 42 -6
49 Rashad Jennings RB NYG 33.381 162.44 203.78 2.5864 6.1217 1 56.5 7.5
50 Michael Floyd WR ARI 33.275 147.75 184.7 5.5557 5.3019 1 47.8 -2.2
51 C.J. Spiller RB BUF 32.841 161.9 204.62 5.2945 6.0785 1 45.8 -5.2
52 Kelvin Benjamin WR CAR 30.101 144.58 186.38 4.8893 6.7775 1 119.9 67.9
53 Matthew Stafford QB DET 28.98 322.41 347.64 5.6729 4.2354 1 47.8
Is there a fluky projection out there or what? I also found it odd that Brandon Marshall had the 2nd highest value while Calvin Johnson sat at 5.
Thanks for catching this, Tim. I isolated the problem (it’s with EDS). I’m working on the fix, should be ready soon.
Just fixed it–data should be good now. Good catch!
-Isaac
Got another auction draft going in a few hours, just discovered your app. Looks awesome. One thing I noticed was that once you get past the starters, there doesn’t seem to be any recommendation for backups. Am I missing it or is that something for next year?
Spend your cap on starters. For backups, use the projections and sleepers apps. Good luck in your draft!
This is great! I started to write a program to do this and found your site and it’s going to save me a lot of work. I’ve started to get into weekly auction drafts this year. Any chance you could add an option to use the weekly projections from each site?
Hey Nate,
Great suggestion, weekly projections are on our to-do list. Hopefully we can do this next season.
Cheers!
-Isaac
Thanks for the apps, Isaac!
I used them in my league and ended up in second place, losing in the finals by only 1.5 pts. Such is life.
Anyhow, is there any chance you’ll release the Shiny code you used to create the apps on your GitHub account? Last season I was worried because we drafted in a bar with spotty wifi coverage, and I didn’t want to be left up a creek in case my connection dropped. I’m trying to mitigate any potential issues this season, and with the scripts installed locally, the last piece of the puzzle is getting the apps locally as well.
No harm if you’d rather keep the apps to yourself — you and the team have clearly done a lot of work on this site, and I appreciate it all the same. Just thought I’d ask.
Hey Adam,
Tough loss, but well done! As Cubs fans say, “There’s Always Next Year.”
Glad you found the apps useful. The code for the apps isn’t public but all the functions it uses (calculating risk, projections, etc.) are public on GitHub, so you could always use the R scripts locally if you’re familiar with R. We are looking into ways to create a standalone version of the apps that you can run locally on your computer without an internet connection. One of the reasons we prefer to use the internet version, though, is so that people can always have the most up-to-date data file. We definitely want to gauge users’ interest on this, so any thoughts or ideas would be helpful.
Thanks for your interest!
-Isaac
Totally understandable. The web version as you have it is fine, really. I can use the data locally as a backup. I’ll make sure to do a pull of the Git repo before the draft weekend.
On the app’s side, one thing you might want to consider: The PPR input box currently only accepts integer values. Our league uses 0.5 PPR, so the closest I could come was 1. It might be worthwhile to accept decimal values as well.
Keep up the good work!
Even though it turns red, it still should accept and calculate decimal values for PPR leagues. Do you get an error?
Aah, good question. Thinking back, it could’ve been that I retreated to integers once I saw the input’s border turn red, without actually trying to run the app with the seemingly invalid value.
I think I might’ve assumed it was disregarding the value, going back to 0 PPR instead of actually calculating the value. I’ll keep that in mind for next time!
Would it be possible to incorporate a running chart in the draft program which shows the top players projected points at each position? As a player gets drafted, he would drop off the chart.
Basically, one chart showing the next 12-15 players available at each position with the charts set up similar to what you have on the following page:
http://fantasyfootballanalytics.net/2013/07/expected-points-by-position-rank-in-fantasy-football.html
I was able to do it by downloading your snake draft data, but currently it is very clunky for me to make it remove players as they are drafted and add the next available player in line. Have a visual chart really helps when looking at what is likely to be available by the next round of the draft. I’d be happy to send the excel program if it would help explain what I am talking about.
I think I explained what I think would be helpful poorly. A line chart plotting Points (or VOR) vs the next 10-15 players available at each position in the draft would be helpful. There would be lines for QB, RB, WR, and TE showing the nest 10-15 best players available. As a player is drafted, his value drops of the chart and a new player is added. Having a visual line chart would make it easier to spot where large drop offs in talent/position occur, and it allows you to better estimate who might be available at the various positions when the draft comes back. If I have historical data showing 2-4 QBs are taken by the end of round two, it makes a quick visual on how many QB points might be available at my next pick. I can compare that slope to my estimated number or WR, TE, or RBs and make the best decision.
Hi Dan,
Good suggestion, I’ll put this on the to-do list. We’re currently working on the auction tool to get it ready for draft season. In the meantime, you can easily hide/exclude players by clicking their name. You can get a season of where large dropoffs occur with the “dropoff column”.
Thanks!
-Isaac
Is there a way to see the actual averaged projected stats as a list, instead of just the points and rankings calculated from the stats?
Hi Larry,
This is on our to-do list. We are currently focusing on getting the Auction tool ready for draft season.
-Isaac
Hey Isaac,
Amazing stuff. Thank you!
I’ve been toying around with the Projections APP trying to figure out the most accurate way to set the VOR Baseline for my keeper league. (14 tm PPR: 2QB, 2RB, 3WR, 1TE, 1DEF). In an earlier post, I read that I should set the baseline to the number of players at each position drafted w/in the top 140 (for my league size).
So in looking at past years, if by pick 141, 33 QBs have been drafted (we skew QBs heavily and back-ups are always drafted in case of injury) i set the VOR baseline at 33?…64 for WRs and so on….
Am I correct in my assumption?
thanks
Hi Adriano,
That’s a perfectly reasonable way to go about it. There are other ways you could calculate VOR baselines, depending on your goals:
http://fantasyfootballanalytics.net/2013/04/win-your-snake-draft-calculating-value.html
Hope that helps,
Isaac
Thanks Isaac.
I’ve been mixing up the VOR baselines per position and comparing them to past drafts to attempting to mirror #’s per position drafted 1-120. The other methods – average, worst and man games – weren’t nearly as reflective of our league’s values.
Excellent work!….if the app could save data settings that would be a great bonus, the App’s live draft functionality is perfect.
Thanks, Adriano. Yes, saving settings is on our to-do list.
Thanks,
Isaac
Hi Isaac,
I’m about to start my second season of ffb and just discovered your app today. This thing is amazing! Every stat I wanna see and all very adjustable in one spot!
I do have one suggestion that is very simple and something I can do myself, but I thought I would mention it here in case there are others that might also find this metric useful.
I would like to see the difference between a player’s projected point ceiling and floor. In other words, the player’s point range as a single value. I like to look at this value as a very quick and rudimentary way of gauging a players projected volatility as I draft. (Of course, this would ignore injury risk and assume the player remains healthy like all other point projections). A lower value in this point range column would indicate more agreement among analysts regarding a players projected points on the season. I find this measurement most useful in mid to late rounds of a draft when I’m looking for backups or sleepers and comparing their upside.
The scale of these ranges would be small (5~20 points) which may not mean much to others. This may be such a small and insignificant measurement that most people wont look at so there is really no need to adjust such a great app.
Thank you again for this piece of fantasy drafting gold!
Brandon
Hi Brandon,
Good suggestion, I’ll add that to our to-do list. We modified the calculation of ceiling and floor to represent the 10th and 90th percentiles of projected points across analysts, which makes the spread larger for many players.
Thanks,
Isaac
Hey Isaac,
This is awesome! Just wanted to know if you are planning to update this tool with 2015 data?
I should have said, specifically the Auction Draft tool. I tried using it but it looks like it still has 2014 data.
Hi Jeremiah,
Yes, we’re working on updating it now.
Thanks,
Isaac
Hey Issac,
As always, awesome site! One question – the highest ceiling lineup optimizer gives dozens of positions instead of the standard 9 for a lineup. Is this an error or do I need to configure the selections differently?
Thanks!
Works for me. What settings are you using?
Help me…I don’t have access to the 2018 season VOR estimate anymore and my draft is monday !!! The Snake optimizer app only shows me the week 1 projection.
https://fantasyfootballanalytics.net/2018/07/2018-fantasy-football-projections.html#comment-23323
Hi Isaac!
I’m a grad student at SMU, doing a group project for a database design class. My group and I are trying to create a sort of “fantasy league” with NCAA basketball data using Shiny. This is a SQL beginners class, and we’ve been encouraged to look for preexisting templates to help us. Do you have a template for this app that you’d be willing to share?
Best,
Elaina