This forum is archived and no longer active. You can visit us in our Discord Server here!

Replied To: Help needed

1

We recommended that you build lineups with as much of our own input/research as possible. Change projections, like players, change the advanced settings.

I'm not sure why you are having difficulty but my hunch is that your player pool has too many expensive players and not enough value guys to balance it out. Add some cheaper players to your pool and see if that helps you.

No promises, we are looking into this.

Once the starting lineup for a team is announced we remove all inactive players on that team from the player pool. Our projections update throughout the day to reflect all the new data that comes in.

First, RG is not a casino and DFS and is not gambling. Second, we don't have a specific guidelines of what can and can not be posted on our site. We reserve the right to remove content posted to our site that we feel does not belong in our forums.

Oreo or Taicheeze,

Can you go into more detail on the randomness setting please.


The randomness setting adds more diversity and variance into your lineups. Your lineups become less optimal but they tend to be a bit more different from your other optimal lineups. Combining this setting with our unique player setting and exposures results in much more diverse lineups. The higher you go on randomness and unique players (and lower on exposures) and you will see more and more different lineups. However, at the same time lineup quality will decrease rapidly if you go to the extremes (low exposure, 100% randomness, 7+ unique players). To combat this, my best tip is to set a minimum salary floor(95% of salary cap) and remove all players you are not interested in playing from the player pool.

What randomness setting do you use Oreo? I see you at the top often...


Depends on how much confidence I have in the specific day. I often go up to 100%.

Tai,

So when setting up lineups, do you recommend leaving it at 100% with 1 unique and 2%? Seems like 100% gives me a lot of 100%'ers if they're projected highly. Problem is, I dont want to be over-exposed to one single guy. Boban is going to do that to me today....follow! Also, what are the thumbs for? Up and down.

Thanks again,


I suggest setting your max exposures to numbers you are comfortable with for each player. The higher the exposure to a single player the greater the risk (and potential for profit). There is no set rule for setting exposures. Personally, I would set at 3 or 4 unique with a max of 50% exposure to any given player I was not certain about. Thumbs up increase the projection when calculating. Its a quick way to increase it by 8% for each thumbs up.

We don't have a current way for you to do this. The best suggestion that I have is for you to decide which types of stacks you want and request a smaller amount of lineups for each type. I know this is not perfect but it's what I have done in the past.

Putting them in groups is a great idea. The next step would be to use "each" exposure in advanced settings and then setting exposure limits for each of your pitchers. Also turning up randomness may give you further diversity. Finally, adjust the projections. Lower projections will make those players show up with lower frequency.

As far as we know he is playing. But this is a hectic time of the year, we don't expect to be able to catch every single "injured" player. Some playoff teams will definitely be randomly resting players in preparation for the post season.

Under advanced settings we have a setting that changes the exposure settings. The default setting is "group" exposure which will behave as you describe as FC will try to make the highest projected lineups possible. However, if you switch to "each" exposure your players will be more spread out throughout your 100 lineups. Give it a try and let us know if this works for you.

Replied To: What time

1

What time of the day do projections come out. Thinking about upgrading, but would need to know this.


Initial projections are posted by 1AM EST each day. The projections are then updated throughout the day.

Sorry if this has already been answered, but what do the grey outlines around certain players mean?


I think you are referring to players thats are unselected and therefore not included in the calculation process.

Its right next to the upload player pool option on the selection bar.

Replied To: EPL/UCL?

1

Eventually, but its not something we are working on at the moment.

We just added a new setting that allows you to randomize projections in between lineup generation. There are 2 main benefits for using this setting. An easy way to think about how this setting works: if randomness is set to 3% the lineups generated will be approximately within 3% of optimal with the given constraints.

1. Increased randomness helps with getting increased variance between lineups.
2. The speed of the calculation increases as the randomness setting is turned up. This is especially helpful with DK NBA and decreasing run time when generating 50+ lineups.

Let us know what you think!

Best,
Jeff (00oreo00)


We are giving away a free month of Fantasy Cruncher EVERYDAY until December! To win, just use our logo on DK or FD. The highest single GPP cash of the day that is reported to our Twitter account (@FantasyCruncher) will win the month. Please include a link and a screenshot to verify that you our running our logo. Min $500 GPP cash to qualify.

Change your DK logo here.
Change your FD logo here.

Here is the image we would like you to use:


Good luck and happy grinding!

We are working on an GPP specific addition to FC that will address this.

A few tips:

1. Limit the size of your player pool to less than 50 players.

2. Reduce the number of constraints (1 unique is fastest, 100% exposure).

3. Firefox seems to work faster than Chrome, although we do not have a extension for FF as of right now.

Not at the moment, minimum exposure is far trickier than maximum exposure. We are thinking about ways this could be done though. Give us some time.

Not at the moment, minimum exposure is far trickier than maximum exposure. We are thinking about ways this could be done though. Give us some time.

We will look into this, but at the moment we are limited by the availability of contests on the site.

Currently we load the next day at 9pm PST for NBA. We might be able to roll this out 1hr earlier.

We are glad you are enjoying it.

Here are a few good articles we wrote to help you get more out of FC.

https://www.fantasycruncher.com/articles/ties-thoughts-exposure-settings-and-its-pitfalls/

https://www.fantasycruncher.com/articles/fantasy-cruncher-best-practices/

Hi Everyone,

We are running this promo again with a new spin. If you use our logo and win a GPP with at least 5000 entrants, we will send you a brand smacking new FC T-Shirt. Also, at the end of the current NFL and NBA season we will be awarding a free lifetime membership to the user with the highest single tournament cash for each sport.

Here is the logo we would like you to use on the DFS sites:



If you win a GPP please post a screenshot and link in this thread showing the logo, as well as your place in the standings. Also, please contact customer service with your shirt size and address.

Thanks and Happy Crunching!

- Jeff

Sorry about this, I will be posting them earlier. Its weird that DK cancels these contests. They are free!

There is no publically-accessible FanDuel API, is there?


nope

he got helped off the court friday and did not return. (knee)
that's probably why cruncher has him projected at zero points. (i don't work for cruncher so i'm not sure if that's why)
he's listed as probable today from las vegas.
type him into the search box and change his projected points and you should be ok.


Thats pretty much why. We made the change when he was confirmed active.

The biggest problem here is actually Fanduels policy towards editing lineups. They don't allow it. If that policy changes we will work towards this. Its definitely something we want to do.

Yes we pull player pools directly from the sites. By default we hide players with projections below 3. If you use player manager you can name any active player and give them a projection.

Hello. I was hoping to use your service to automate what I would normally have to do manually, but I see there are often issues when trying to do this. Either this is a programming error on your side, or simply, your software can not perform this task.

I've seen this happen many times in the past 9 days, but I will only post one example to show this The goal here is to generate every possible lineup, given the constraints I have selected. To be clear, in this case, all I am doing is limiting the player pool. ALL other setting are at default. I do NOT have any of my own projections inputted. I have set it to generate 100 lineups, but all is throws back to me is 8 distinct lineups, which is far from being correct. Again, I have and can duplicate this with other games, but will only use this example as it seems to be the most obvious when showing this problem. Here is the setup and result:

1) Chose only these 2 games: NE vs IND and NYG vs PHI.
2) Chose only 1 player at each position from each team. In all cases, the most expensive player was chosen at each position, from each team. Given that, using the NYG as an example,...Beckham and Jennings was chosen. In the end, each team has 6 players in the pool, for a total of 24 players. "Players" in this case obviously includes DEF as well.
3) Lock Brady as the only QB.

So, with this setup, given the small player pool and taking into account the salary cap, we should have 30-60 possible combinations. Instead, this only generates 8 possible lineups. Not a single NYG player was used in any of the 8 generated lineups. Very strange. Obviously, the salary cap is not an issue.

What's also very strange, if I make only 1 change,...replace Beckham with Randle at WR fpr the NYG, it generates 38 lineups, with every Giants player sprinkled in at some point in those 38 lineups. If I then swap Beckham and Randle again, I once again only get 8 lineups.

To be clear, I have repeated this test of multiple computers, and multiple browsers. I have rebooted, dumped browser cache, logged out of FC and then back in, etc, etc. Also,...I have changed the drop down to each setting,...FC Projections, Highest Celing, etc, but it makes no difference. Thus, this is not an issue on my side.

Once again, I can repeat this with other games and other players, but I am using these games and players as it's the most easily replicated, and the most obvious. Not being able to generate all lineups from such a small player pool is a major problem. I only signed up to be able to do such things to save myself from doing it manually, but as of now, it seems your software can't do this each and every time.

Any help or an explanation would be appreciated.

Thanks.


It is very likely that when you have both Brady and OBJ there are very few combinations that can be made given the current settings. Are you able to make a 9th unique lineup by hand using your current player pool that does not break the rules?

I asked an almost identical question in the "Everything else" section last night ... ... ...

I guess based on this reply I would ask ... ... ...

Fair enough, and I signed up without any intention of using the projections ... but I guessI'm more wondering, especially with you guys downplaying them, if I should factor them in at all? I often will crowd-source or aggregate projections, but I can't get a feel for whether these are useful. I can LOOK at football projections and be like "these are junk" or "these are solid" but I can't do that for NHL, because I know nothing about NHL.

So - I mean you can be honest! Are the projections solid? What are their weaknesses and what are their strengths? So like for baseball I have a set that is A+ for pitchers but terrible for batters, another set that is the opposite, and another set that best factors in all of the matchups and park factors, etc. - and between the three I can piece together I pretty sharp set. I'm used to that. I didn't sign up for the projections. BUT. They are there.

I guess you mentioned a stat database - but for NHL all I see is a line number and a final projection ... maybe ceiling but in any case ... Is there a stats database for NHL that we can view?


I would say our projections are very good if not better than most sources on the internet. We take pride in this and want them to be as good as possible. IMO we tend to be better at sports with less variance. So ranking sports in terms of variance I would go NBA > NFL > MLB > NHL. With golf being more or less a crapshoot because of cuts and the bonuses involved with the players final place.

This actually corresponds quite well with DFS sports I play. I have my highest DFS volume during NBA season, second highest with NFL, then MLB and I don't play very much hockey. I avoid DFS PGA/MMA/Nascar.

The models we use vary from sport to sport and we make changes over time. We use a ton of factors such as historical performance and various advanced metrics. We don't factor weather into any of our models. Some of the sports require more of a human touch as models won't cut it on their own.

With all this being said, we don't feel like you should be subscribing to FantasyCruncher for the projections alone. There are many sites that offer projections as good as ours, many of them being free. What you are really buying is access to our optimizer, export tools, and stat database. There is enough information on our site for you to build better lineups for all types of DFS contests in significantly less time than you could without us. The value of FC varies from user to user, if you are a lower volume DFS player you may find that FC is not the right site for you.

Its still a work in progress. If you have any suggestion please let us know here.

I like it! We may make a section entitled "Blank Optimizers". Not sure how quickly we out it out, but I'm adding it to the to do list.

Replied To: Chat

1

We don't have a chat onsite. But alot of our users have joined a small little group on Slack. https://slack-dfsports.herokuapp.com/

None of the projections would have changed in the morning. I already has Stat, Melo, Ilyasova as assumed playing so none of the numbers would have changed on our end.

I don't know how you have the time to update projections and enter all the contests you play in.

You should ask for a raise.


To be honest I need to update the projections to build my own lineups.

Projections are updated randomly. Pretty much whenever I have time and news breaks. We will be adding another person to help with the frequency.

Replied To: Victiv NBA

1

This is definitely on our long to do list

Replied To: Lineup Rewind

1

Thanks CeeGee, this is very helpful. I had James Harden and Westbrook locked yesterday and i completely missed the min as James Harden was about 25 pts short of his projection. I feel like his performance slumped due to the back to back game and with overtime too, there might be some fatigue. anyways, lesson learned. i'm trying to get a better hang of which players to lock. i think even a westbrook/davis lock would have done much better.



I went with the Westbrook/Davis lock and just had an OK day. I also took a risk on Kevin Durant and him rolling his ankle really limited my upside.

Replied To: test video

1

draftkings.com/r/00oreo00

www.draftkings.com/r/00oreo00


www.draftkings.com/r/00oreo00

Its tough to do this. The way we have it set up you can edit the projection. If becomes difficult if you are able to edit both the projection and the minute total. It becomes even more complicated if you are allowed to edit the fantasy point per min too. We want the site to be simple to use, and don't plan on adding this feature right away. We may introduce an advanced optimizer in the future.

Replied To: projections

1

NBA projections are updated once the night before. Going forward we will try our best to update one more time around 12pm EST and finally 1hr before NBA lock.

Thanks for the feedback and we are glad you are enjoying the site so far.

@oreo,

In response to #1, the Update Time... Do you have a general time that you like to make your last update each Week for NFL?

Also, generally, do you have a time that you like to have the NBA day fully updated?


NBA will typically be updated 1hr before lock.

News is not as frequent for NFL, so this will be very sporadic. So Tuesday, and once during the weekend.

Replied To: donald sloan

1

Hey guys. This is a interesting discussion. I just posted my thoughts in a "best practices" post.

http://www.fantasycruncher.com/forum/viewtopic.php?pid=119#p119

1. Be aware of the last update time. Our projections include a time stamp indicating the last update time. News can have a large effect on the projections. If a significant event happened after the projections were last updated then you should try to account for this event on your own. Accurately accounting for last minute injury plays is where alot of the edge comes from in daily fantasy sports.

2. Please don't blindly use our optimal lineup. Vet every single player in the lineup. Make sure they are healthy. Also check their last few games and make sure they are not slumping. If something seems off then move on to the next player/lineup. Trust your gut, a tool can only do so much.

3. Don't over expose yourself to a single game. Having too many players in a single game really puts you in a bad spot if the game gets blown out. We are working on a way for you to easily limit exposure in the Cruncher.

4. Are you trying to cash in a tournament? Think twice about how you approach lineup construction. Fantasy Cruncher lineups tend to include many popular plays. Using popular plays makes it really difficult to differentiate yourself from your tournament competition and finish with a top lineup.

5. TWEAK! One of the best parts of our site is that we give you the ability to tweak all of the projections. Change individual projections to a number you are comfortable with. This is the best way to insure that you are making YOUR optimal lineup.

6. Some research is required. Many factors can not be accounted for in projections. Things like birthdays, short slumps, hot streaks, revenge games, and minor injuries can not possibly be included in our model but can have huge effects on player production. If you think a player has extra incentive to perform then lock him in or bump him up.

Good luck!

Replied To: Lineup Rewind

1

Interesting analysis! The projected lineup will not change in Lineup Rewind. BUT what might change our floor/cieling numbers. These numbers are currently ties into the main database. If a player has a really good game his ceiling will increase. If a player has an especially bad game his floor may decrease.

Posted Topic: Glossary

1

Fantasy Cruncher Specific

Consistency: a value that takes into account a players standard deviation and average. A low consistency number indicates that a player is less reliable and can be very streaky.

Ceiling: an estimation of a players upside on a good night.

Floor: an estimation of the lowest score a player is expected to score on a bad night.

Projection: estimated fantasy score. Fantasy Cruncher takes into account previous performance and matchup difficulty to generate daily projections.

Value: the number of points you expect per $1000 spent. If a player has a projected value of 5.0 and cost $10,000 you can expect that player to score 50 points.



DFS Glossary

50/50: A league with more 4 or more players. The payouts are structured in a way that half the player pool wins money. The payouts are flat so all winners receive the same amount of money.

Cash game: these types of contest typically fall in to two categories, 50/50’s and head to heads. In both games every player has a 50% chance to win.

Daily fantasy sports: one day or one week fantasy sports.

Double-ups: a contest in which winners receive double their entry fee. Double-ups typically pay just under half of the player pool.

DFS: an acronym for daily fantasy sports.

Expected value: estimated return. Smart players look for positive expected value opportunities such as playing in a contest with overlay.

Fade: avoiding a popular play.

Freeroll: a free DFS competition. Several of the sites offer these on a regular basis. Freerolls usually include small cash prizes.

GPP: guaranteed prize pool or a tournament style contest. Daily fantasy sports sites guarantee the payout and fix the contest size. The prizes will be paid out regardless of how many people join the contest.

Late-Swap: some sites allow you to make changes up until a player’s game is scheduled to start. Changing out a player after the contest begins refers to a late swap.

Lineup: The set of players you enter into a DFS league.

Live final: One of the largest GPP’s a site offers, entry is obtained by winning a qualifier. The event takes place in a fun location and all contestants are invited to join. The winner of this GPP typically wins a very large cash prize.

Lock: The time that a player or lineup can no longer be changed.

Overlay: Occasionally a DFS site will run a GPP that does not fill. If the site pays out more money than they collect the contest had overlay. Overlay helps players because every dollar invested has positive expected value.

Qualifier: a tournament in which the top prize is entry into a live final at the end of the season.

ROI: return on investment, this is a measurement of a particular players success of failure. A positive ROI indicates a player who is winning money.

Rake: the portion of entry fees that the site keeps and does not pay back to the player pool.


Reverse Fade: Intentionally playing a popular play that stupid people will skip out on

Satellite: a contest in which the top prize is a ticket into a more expensive contest

Stack: playing multiple players on the same team in the same lineup.

Replied To: test video

1

**bold** bold

Replied To: donald sloan

1

Sorry about that. The model we use requires having accurate depth charts. The current depth chart we are using had Sloan as the backup PG. This was a weird situation in Indiana where a player went from averaging 27 mins to 1 min over the course of just 3 games with no injury report. I'm making the change to remove him from the active roster tomorrow if he does not get play time today.

I love what you guys are doing. I think it's fantastic you are sharing your hard work with others. I've been using the tools for a little over 2 weeks now, and I've been very impressed. I am curious about something though. The tool updates what seems like 2 or 3 times per day. I've been waiting to set my lineup much later in the day when possible. There have been times when I set the lineup early in the morning, check it again that evening and get entirely different results. I'm going to assume that waiting until later in the day to finalize my lineup is the best way to go? Or not?



Thanks. NBA DFS is extremely dependent on news. So I try to update the projections when something big happens. I personally don't start building lineups until 1hr before the lineups lock.

Not at the moment, but this is something we will incorporate into a future update.

Hi TheOtherDave" Love this site!!!

Can u please explain to me the difference between the best "projected", "ceiling" and "floor" line ups?


Right now we have it setup so you can make a lineup with the highest possible aggregate number. If you optimize by floor, you are only looking at the number we provide for floor and we try to make the highest total floor lineup.

Same with ceiling, the optimizer ignores the projection and only uses the ceiling numbers.

Personally, I only optimize using the projection. In my experience these tend to do better on average.

Right now everything is based on defense against position. So we might now something like Memphis allows 20% less points but 10% more free throws against scoring guards. All of this is used to calculate an accurate prediction for each player.

Of course if someone like Tony Alan were to sit out then our stat multipliers would be off so we also do quite a bit of manually adjusting to make sure our projections are accurate every day.

Posted Topic: Bug reports

1

If you find any bugs please let us know here. We are still in beta and this type of feedback is very much appreciated.