The Official Food Club Topic

[jcpet]

Decided to close my eyes and play 4 arenas today too. No double safeties covered either...

 

Scurvy x Goob x Bonnie x Ned

Fran x Goob x Bonnie x (Fed, Fairfax)

Fran x Scurvy x Buck x Bonnie

Fran x Scurvy x Buck x (Tailhook, Peg Leg)

Fran x Scurvy x Goob x Bonnie

Fran x Scurvy x Goob x Bonnie x (HC minus Squire)

[Bright ^_^]

So, I've started playing food club recently, and I've been following Hugo's bets. But today Hugo didn't update his debts until just recently ---I only had enough time to place 8 of the 10 bets. I think I missed out on one of the boosters and one of the safeties. Apparently he's going to be traveling. Maybe now I'll have to learn to bet for myself. :( Maybe I can figure out how to follow his system to make my own bets? But after the basic bets and the safties, how do you know which booster bets to pick?

[siniri]

Here's the easiest way: look at the total starting probabilities for each arena (1/odds), using the Daqtools link shared earlier to get the true starting values. The total of all probabilities in an arena should add up to 100% (because somebody has to win the arena, if you bet on all of them, you should have a 100% chance of winning). But Food Club is a little weird. Odds are always rounded up to the nearest whole number, with the highest odds ever given 13:1. So often the probabilities won't add up to 100%. The ones that are below 100% mean that the 2:1 pirate in that arena actually has a better chance of winning; these become your "positive arenas." All neutral and/or negative arenas become your "booster" arenas. The closer the sum of the probabilities is to 1, the higher expected value ("average" winning over the long term) it will tend to have, so usually those are the better arenas to choose. (This is because an arena with a "total probability" of 1.4 would mean that across the 4 pirates, their actual true probabilities are significantly lower -- about 10% each -- in order for them to add up to 1. So one paying at 2:1 would only have a true odds of 2.5:1; it would only win 40% of the time, so in the long run, you'd only make 80% of what you bet. So why would you bet on these negatives at all? Because their negative is canceled out by the positive EV from the base arenas. But you want to choose the least negative to do the least damage to your EV.)

 

A couple of notes:

1) Because 13:1 is the highest possible odds, these pirates could have an actual odds of 12.0001:1 to infinity:1 (i.e., no chance of winning at all). I tend to assume that the odds are 25:1, unless Hugo says that they have a good shot (in which case I leave it as 13:1). I sometimes leave these pirates out to place more bets, but there's a benefit to betting an entire arena, too. Just make sure they really started at 13:1 (10:1 and beyond almost always move to 13:1 fairly quickly, and these will tend to add to the expected value -- see below).

 

2) The odds often change, tending to grow longer for some of the middle and longer pirates in some of the arenas (especially the positive arenas). Obviously, if the pirate has a true 4:1 (i.e., 3.01:1 to 4.00:1) odds of winning, but is now paying out at 5:1 or even 7:1 odds, betting on him would increase the expected value of your bet. Sometimes I'll bet just on the ones that have moved in a negative arena, and sometimes I'll bet the whole arena; it just depends on what other possible bets there are and how much worse/better they are. Sometimes pirates can also move the other way (4:1 to 3:1, or 3:1 to 2:1) -- I find that these tend to have negative expected value, so I tend to avoid them like the plague, no matter what their odds of winning. In the long run, they'll lose you NP. And that's not why I play Food Club.

 

tl;dr: To pick booster arenas, add up the starting probabilities (1/odds) for all pirates in the arena, selecting the ones with total probability closest to 1, and decide whether to include 13:1 pirates, which often are longer shots than their odds indicate. If boosters are similar, pick the ones that have more pirates with increased odds over the course of the day, or pick the ones that give you a 10-bet system that makes sense to you.

 

Side note: Hugo likes to pick booster arenas that he thinks have a bigger chance of an upset. These tend to be ones with only a single 2:1 pirate and large negative (sum probability way over 1). (A 2:1 pirate could have a true probability anywhere from 34% to 88%; a 3:1 pirate can only have a true probability from 26% to 33%; a 4:1 from 21% to 25%; etc. So the farther the arena is from 1, the more likely the 2:1 is actually pretty close to a 3:1 (because the others just don't have much room to absorb the "extra" probability), meaning the other pirates in that arena have close to a 2/3 chance of upsetting combined.) This can be a good strategy compared to picking an arena with two 2:1 pirates (which combined have at least a 68% chance of winning, which leaves little chance for an upset).

 

Anyway, hope you managed to get the Ned safety in. 24:10 for me today. Not the 208:10 I was obviously hoping for (but not really expecting), nor the 48:10 I could have gotten with the double safety, but I'm happy. Saving it for now in hopes of a trophy upgrade.

[Bright ^_^]

tl;dr: To pick booster arenas, add up the starting probabilities (1/odds) for all pirates in the arena, selecting the ones with total probability closest to 1, and decide whether to include 13:1 pirates, which often are longer shots than their odds indicate. If boosters are similar, pick the ones that have more pirates with increased odds over the course of the day, or pick the ones that give you a 10-bet system that makes sense to you.

 

I didn't really understand much of what you said. :/ But I'm hoping that once I start trying to make my own bets it will start to make sense. And I missed the Ned safety.

[siniri]

If you've never studied probability and statistics before, you may want to find a basic introduction (you really only need the very basics to understand Food Club, no more than a chapter or so -- I'm sure you could find a free resource online that explains these simply, usually involving dice or coin flips -- you also might want to find a simple explanation about the difference between odds and probabilities). Once you've read that, hopefully my explanation will make a bit more sense.

 

Today has 3 positive arenas. Hugo ignores one of them, but Scurvy still has a 54% chance of winning that arena, so in the long run, it's better to include him.

 

My bets (based mostly on highest EV, with one adjustment: forgoing the base bet for the Squire safety):

Goo x Ed x Scurvy x (Fran, Peg, Fed)

Goo x (Lucky, Squire) x Scurvy

Goo x Ed x (Black, Young, Orvinn)

Goo x Lucky x (Young, Orvinn)*

 

Overall chance of winning something ~74%. Overall EV=2.31:1. Using Hugo's system, but replacing SW with Fran, Peg, Fed from HH and the base bet with the Goo x Squire safety has ~86% chance of winning something and EV of 2.22:1. Using Hugo's bets as he made them has ~78% chance of winning and EV of 2.04:1 (yes, you read that right -- worse probability and worse EV than if you made the two changes I suggest).

 

*these each have ~2.8% chance of winning and EV=2.35:1; you could substitute for the base bet of Goo x Ed x Scurvy w/ a 30% chance of winning and EV=2.43:1, plus either Goo x Ed x Scurvy x Puffo with a <=2.3% probability and EV=1.26:1 to 2.43:1 (to round out the HH arena), or the Ned x Ed x Scurvy safety that Hugo thinks is necessary, but has a probability of <=2.8% and EV=1.45:1; I chose the two 84:1 potential payouts over the more likely 8:1 combined with one with lower EV and similar probability as the 84:1's.

[jcpet]

Playing it safe today. I don't trust Gooblah for some reason, even though he has 85% chance of winning. I'm just trying to get consistent (even if small) wins for the trophy. Been holding since I got that 108:10, and kinda banking on that for the trophy.

 

Goob x Lucky

Goob x SEO x (HC minus Scurvy)

Goob x SEO x Scurvy x (HH minus Puffo)

SEO x (HH minus Puffo)

[siniri]

Yeah, if I were sitting on that, I'd be playing pretty conservatively, too, jcpet. Since I just started holding yesterday, I'm behind, so I'm playing a bit more aggressively.

 

ETA: Should have kept the base bet and ignored the Squire safety; I would have gotten 24:10 instead of 16:10. Ah, well. I like to have at least a 75% chance of winning when I can, and the Squire safety would have paid nicely had it come through.

[Cassia]

Ugh I misbet yesterday by accident, since I was betting while half asleep. The bets that I was GOING to place would have gotten me 20:10, but instead I got 12:10. Oh well, better than nothing!

[jcpet]

20:10 yesterday; getting double or more is always good :D

 

May not have time to bet tomorrow so I'll bet tonight.

Three arena day, all looking relatively safe.

 

Tailhook x Buck x Bonnie

Goob x Bonnie x (Lucky, Crossblades)

Goob x Buck x (Stripey, Sproggie)

Goob x Buck x Bonnie x (Harpoon's Harry)

Tailhook x Buck x Stripey

[siniri]

20:10 yesterday; getting double or more is always good :D

 

May not have time to bet tomorrow so I'll bet tonight.

Three arena day, all looking relatively safe.

 

Tailhook x Buck x Bonnie

Goob x Bonnie x (Lucky, Crossblades, Peg Leg)

Goob x Buck x (Stripey, Sproggie)

Goob x Buck x Bonnie x (Harpoon's Harry)

Tailhook x Buck x Stripey

Isn't that 11 bets? How can I swing an extra one? Or did you leave off someone in HH?

[jcpet]

Isn't that 11 bets? How can I swing an extra one? Or did you leave off someone in HH?

 

Whoops. Originally wanted to do a safety on Peg Leg, but decided on the double safety instead.

[siniri]

My bets are a bit risky today, but they're based on EV. I need to take more risks if I want to upgrade my trophy this month.

 

Goo x Buck x Bonnie x (Black, Ed, Fair, Ned, Puff, Fed) (yep, if there's no upset in either booster, I'm getting bupkis -- but there's only ~15% chance of that happening, so I feel pretty good about my chances)

Tail x Buck x Bonnie

Goo x (Lucky, Cap'n) x Bonnie

Goo x Buck x Stripe

 

I would have loved to include the double-safety on Tail and Stripe (and the Young Sproggie safety, and the base bet, and the last 2 pirates in each booster arena...), but there was no room.

 

ETA: That Bonnie safety paid better (36:10) than I would have gotten without the upset (32:10). Woohoo! I only have 76:30 saved up so far for the trophies (3 days in a row); I know I'll need a lot more luck to get gold, but maybe I can upgrade to silver.

 

9/30 Edit: 2 positive arenas. My bets (going by strictly positive EV ended up with some wonky bets; I wanted the 4 safeties because they had some of the best EV, and then rounding out the TI booster made a lot of sense because the last bet still had decent EV and was most likely, but this left me with 2 extra bets -- I could have gone with 1 in each other booster based on EV, but I wanted an increased chance of winning, so I went with the best 2-bet combo, which still had decent EV -- the two 2:1's in HH):

 

Fran x Buck x (all in TI, Goo, Puff)

Ned x Buck

Fran x Fair

Fran x Ed

Ned x Ed

 

Edit: I wish I hadn't had to rush my bets today, or I would have considered the other double safety today (you know, the one that actually won...) -- it had great EV and ~3.5% chance, so I would have included it and probably the base bet instead of the second arena. Dratsicles.

Edited September 30, 2013 by siniri

[jcpet]

Had to rush my bets today too and actually put in the same ones as you did, siniri, so nothing for me. Hasn't been the greatest week for a trophy run... busted three times in the past seven rounds, and not having high hopes for tomorrow either because there are no positive arenas.

 

 

Edit: Oct 1st

 

Playing two arenas today, Shipwreck and HC, since Fran and Scurvy Dan both have over 50% chance of winning according to daqtools.

 

Scurvy x (Crossblades, Orvinn)

Fran x (Peg Leg, Lucky, Fairfax)

Fran x Scurvy x (Harpoon's Harry)

Orvinn x Lucky

[siniri]

Hidden Cove opened neutral. Unlike Hugo, I'm not going to assume that arenas opened positive when they didn't; instead I'm going to take advantage of the things that moved to eke out some positive EV. Since HC opened neutral but now has positive EV for all but one pirate, it makes sense to me to use it as a booster arena (the other option would be HH, which has more chance of an upset, but lower EV overall -- though even that arena has moved into positive EV overall, barely). So now I need an arena that has only a few pirates with a combined great shot at winning, preferably ones whose odds have moved. A safe play would be TI, with its double 2:1 pirates (83% chance overall that one will win), but they have negative EV. Instead, I'm homing in on SW, with its two pirates with now-positive EV (Cap'n and Orvinn), with a couple of safeties on Fran since he does have a 48% chance of winning (you could use Fran and Orvinn x the boosters, and do 2 long-shot bets using Cap'n instead, for a little more security).

 

In summary, my bets:

Orvinn x (all in HC)

Cap'n x (all in HC)

Fran x (Lucky, Peg)

Overall EV=1.6:1, 65% chance of winning something (w/ 17% chance of winning only 8:10)

 

Alternative 1 (more conservative):

Fran x (all in HC)

Orvinn x (all in HC)

Cap'n x (Lucky, Peg)

Overall EV=1.5, 86.5% chance of winning something (with 24% of that being a chance of winning only 4:10)

 

Alternative 2 (super-conservative):

Goo x (all in HC)

Buck x (all in HC)

Ed x (Scurvy, Lucky)

Overall EV=1.1, 90% chance of winning something (but with 41.5% chance of winning only 4:10)

 

As you can see, the EV for Alt. 1 is fairly close to the EV for my bets, and much higher than for Alt. 2; Alt. 1 is also the only choice with >50% chance of positive earnings. Really, no one should do Alt. 2. Do alt. 1 if you don't want much risk. Or you can use my bets for a slightly higher EV overall, but with more risk.

 

ETA: Won 24:10, woohoo! I'm sitting on a bit over 800k. Had I started collecting when I could have, I would currently have a good shot at silver. As it is, I've got 3 more days during which I can collect. Gold will require a lot of luck, but silver might be possible (still requires decent luck, though).

[jcpet]

Good thing I chose the correct double safety today. 24:10 for me as well. Sitting in third, with two hours to go... I hope I don't get bumped, but it's safe to say I at least got silver!

 

 

Bets for Round 5283:

 

One arena opens positive, but I play two arenas because I don't like playing just one. Buck has 54.2% chance of winning, so I'm going with HC and HH.

 

Buck x (Ned, Sproggie)

Scurvy x (Stripey, Lucky)

Buck x Scurvy x (Treasure Island minus Squire, Shipwreck minus Blackbeard)

 

No double safeties covered today, so I'll bust if that happens. Also hoping Blackbeard won't pull anything sneaky--he does have the highest strength in Shipwreck. I could've gone with Lagoon instead, but Shipwreck gave a higher expected value.

 

If you want to include double safeties, I would take out Shipwreck and do safeties for Sproggie x Stripey, Sproggie x Lucky, and base bet Buck x Scurvy.

[siniri]

Congrats on the gold, jcpet!

 

I had less than 10 minutes to look at the arenas and place my bets today. I didn't actually calculate EV's (except I knew which would provide the best in my head). Went with the single arena since Buck was not very likely and I was short on time -- and obviously thankful that I did since he lost. I just wish Scurvy had lost, too!

 

My bets:

Scurvy x (Fed, Bonnie, Tail, Buck, Stripe, Lucky)

Ned x (Buck, Lucky)

Young x (Buck, Lucky)

 

So I ended up with a 12:10. If only Ned had won, I'd have had a 52:10 instead. I've got 2 days left to get into trophy contention; if I cashed in now I'd be just shy of repeating on bronze. Two positive arenas tomorrow, but Lagoon is a bit risky. I'll probably play both, anyway, with quite a few safeties. At this point, I just need wins -- even breaking even counts as an 80k+ "win" for me, which could be enough for a bronze. Of course, I'm hoping to find (and win) some big bets in there, too! Gold is a stretch, but silver could be doable with a bit of luck.

[jcpet]

Thanks for the congrats siniri!

 

8:10 yesterday.

 

My bets are the same as Hugo's today:

 

Goob x (Fed, Tailhook)

Goob x Buck x (Shipwreck, Treasure Island)

 

If you want to include Lucky as a safety, I'd take out Treasure Island and use Hidden Cove minus Stuff-A-Roo as a booster arena.

[siniri]

I considered doing Hugo's bets, but that Lucky safety has a 7% chance, and I want to maximize my chance of winning something. I considered using TI and HC (minus Stuff) to have a higher guaranteed minimum, but the difference would only be one place on the score table, really (mid- current bronze range) -- and the difference in the maximum is too high to risk missing out on a big Shipwreck payday (which could potentially put me in the running for silver).

 

So I'm following jcpet's alternative. And rooting hard for Squire, Buck, Goo, and Orvinn.

 

Still kicking myself that I forgot to start hoarding earnings on the 24th; I think I'd have had silver if I had remembered, because I cashed in on some decent-sized wins. Ah, well. If I can't make it this month, there's always next month.

 

(I wish I'd had room for the double safeties -- at least the one on Fed x Lucky.)

 

ETA: Well, had I been less conservative, I'd have gotten another 64.5k (had I gone with TI instead of HC and not done the extra safety). But I'm glad I went with Shipwreck instead of TI (it got me another 32k). I could cash in today and have a very good chance at bronze, but I've got another day to save, so I may as well try for silver. Two positive arenas tomorrow. I could get into the running for gold if I were to win over 750k. I need about 120k for silver (based on current standings, + a bit extra, I'm sure, as others move up), which is more likely. It'll still be hard, but it's worth a shot.

[blueberrykiwi141]

I won 130K in profit today!!

So how do you get on the score table?

You have to hoard your winnings??

That sounds a little weird to me...

[Rebecca~]

Yep, exactly that. :) You don't collect your winnings for eight (?) days, and that cumulative profit is your score. When you go to collect it, the score is 'sent', and if it's high enough, you get on the HST.

[siniri]

Well, this is exciting. Apparently a lot of people started hoarding the day before I did, because the high score table jumped up a ton yesterday. Or maybe they just won huge yesterday. (Note to self: when trying to get on the high score table, do not make "safer" bets with lower EV. Go big or go home; eventually you'll get it, and it's not worth sacrificing daily NP trying for the trophy.)

 

Anyway, can't undo yesterday's stupid betting. Today's booster arenas are all pretty horrible -- HH in particular has the absolute worst negative arbitrage you could get. And having a pirate in Treasure Island move to 2:1 means the only decent booster arena is now not very good. Thankfully the base arenas are fairly safe, at least.

 

When choosing my bets, I always highlight the best possibilities by EV (usually by 100-NP intervals), then reduce down to 10 bets trying to make logical combinations (occasionally adding ones that weren't highlighted to complete an arena, though I never make a bet with negative EV). Today, there were 11 bets with EV over 13k. Removing the base bet leaves me with 10 bets. I don't cover any of the booster arenas fully, but I need there to be upsets to have any chance of getting on the high score table now. And betting by positive EV pays off best in the long run, so even if I get bupkis today, it's all good. I'll get a gold eventually; it's just a matter of time. And now I know my best chance of that is to just follow my system.

 

So, today's bets (I'm not actually making them yet, though, because I don't trust things not to move -- I'd like to bet around 1:30 NST if I can be near a computer at that time -- TI seems pretty unstable to me, and I wouldn't like to bet on someone that moves to 2:1 and have Scurvy move back to 3:1):

 

Goo x Tail x (Lucky, Squire, Fed, Stripe, Young)

(Orvinn, Capn) x Tail

Goo x (Bonnie, Ed, Black)

 

The only bets not enough to get me on the current bronze list by themselves are the ones on Stripe and Young; I need either Fed and/or one of the upsets in Shipwreck (or a safety, which combined actually have ~41% chance of winning). Chance of winning in TI is 33.5%; chance of winning in Shipwreck is 13.5%. Not very good odds. Meh. It's the best I can do with such wonky booster arenas.

 

Note: the only bet with negative EV is the long-shot on Stuff, and it's barely negative, because the base arenas are so safe. Even though Scurvy moved to 2:1, the bet on him is still slightly positive. So you can't really go wrong. Right now, the best booster arena is TI, followed by SW, with HH coming in last.

 

ETA: And Scurvy went back up to 3:1 after Hugo announced his bets. Now I don't know which bet to drop... I suppose it should be the one with the least chance of winning (Squire, 4.4%), but it also has the second highest EV (over 2:1, just behind Fed, who has the second least chance of winning, at 4.8%). Another option is the lowest EV bet, but that's a tie between the Orvinn and Captain safeties (both have a 6.3% chance of winning, slightly less than the Blackbeard safety, which has only a slightly higher EV, too). Decisions, decisions! Guess I just need to keep watching.

[jcpet]

36:10 yesterday.

Today's bets:

Goob x (Bonnie, SEO, Blackbeard)
Tailhook x (Orvinn, Crossblades)
Goob x Tailhook
Goob x Tailhook x (Treasure Island)

Very small range of possible winnings with these bets:

 

64.958% chance of 16:1
18.530% of 26:1
4.868% chance of 48:1

11.643% chance of bust

[siniri]

*sigh* I considered your bets, jcpet, but the most likely scenario wouldn't have quite gotten me into bronze. So I took out the Squire bet, leaving a single bet in Shipwreck. Come on, Lucky, please be super-lucky today! If not, please let there be a single upset in one of the base arenas...

 

Thank goodness the site is finally working better; I was worried I wouldn't be able to make my bets. The site was still slow, but I placed bets in four separate tabs at once, and that helped me get them all in time (with over 10 minutes to spare, even).

 

ETA: ergwqgtjqewr0gotawrjhgewraoghanrogj Arrrghhhhh. Two days in a row I second-guess myself and cost myself major upsets. But somehow my Scurvy bet paid better than it should have, and I might get bronze. Not an upgrade, and I'll have lost more in interest than I'll get for the trophy, but I learned a lot the last couple of days.

[Clumsy rockyroad1]

Round 5286:

Goob x (Fair, Bonnie)

Goob x Fed x (Buck, Stripey, Lucky, SEO, Tail, Fran, Dan)

Goob x Fed

 

Wasn't able to bet the last few days.

[siniri]

Today's bets:

(Fair, Bonnie) x Goo

Fed x Goo x (all SW, all HC)

 

These might be the same as Hugo's? Don't remember, but they're pretty much straight EV (except technically the base bet of Fed x Goo has slightly higher EV than the Scurvy and Buck boosters).

 

P.S. Not surprisingly, given my results the last two days vs. what I should have gotten, I slipped off the high score table (I'm presuming before the NP were handed out -- I don't remember what I finished with yesterday, and I've got some other trophies going). Next month, maybe. From now on I'm not trying, just doing my thing.