Champions that get back in the pool after selling or dying.

PenPen·Jul 1, 2019·5 votes·15,649 views

Ok so I think I discovered something important but I need someone to test this with premades for example.

-When someone buys a champion it decreases the pool of that champion by 1. -When someone sells a champion it increases the pool of that champion by 1.

But here is what I think but should be tested:

-When you sell a 2-star champion does that add 1 of that champion back to the pool or 3? -Does this also apply to when players die

Because what I think is that selling a 2-star champion only adds 1 of that champion back into the pool. Meaning you can deny other players there upgrades by hoarding 2-star champions and selling them later (since you do get full gold back for that).

Ok so then there is a 2nd part to this: -What if everyone stacks up emptying the pool? I believe that this actually works and can easily be abused. If everyone stacks up 1-cost champions you will get a lot of 2-cost champions in your pool. And if then everybody also stacks the 2-costs you will get a lot of 3-costs in the pool etc. etc. But now you're probably thinking if you keep doing this wouldn't it eventually empty out? No since the battles are automatic and also automaticly places units on the board if you dont do it. Which results into players dying and adding those champions back into the pool. But what happens if everyone levels up the champions and then sell them to decrease the pool (if that part acutally works)? Well, that needs to be tested. Which can probably only be done with a full premade team of 8 people.

So if there is a 8-man testing team to test this out can they test this out for me? Also happy to hear what other people discovered about this so far.

Edit: So far I found out that selling champions does put them all back into the pool (so selling a 2-star will put 3 of those back in the pool) I'm still not sure what happens when players die but I have a new theory for people to test. I think there are 2 options which are that all champions of that player go back into the pool or that only the champions on his bench go back into the pool (like i said I'm not certain about this and this needs to be tested). Then there is still the thing about the amount of champions in the pool. Read the comments below for that ;).

40 Comments

PekiCodexJul 1, 20192 votes

I've decided to crunch some numbers, and here's what I've got:

Each Cost has a % chance to appear based on your level. For example at Level 6 there is a 29% chance to receive a Cost 1 champion in the store, 29.5% chance for a C2 champion, 31% for C3, 10% for C4 and 0.5% for C5.

This means, no matter how improbable, there must be at least 5 champions of the same cost available for every player. With 8 players, that means it needs at least 40 champions of the same cost in order to keep these %s

However if there were less than 40 champions in the pool, then the %s would have to skew to reflect this. As such I have calculated how many champions from the cost pool each of the players would need to get in order to get less than 40 champions in that pool. The numbers are as follows:

C1: ~49 C2: ~35 C3: ~27 C4: ~10 C5: ~3

ThingamajigJul 1, 20192 votes

drop % are known https://i.imgur.com/ayhR5Qc.png https://i.imgur.com/rH2qRJL.png

https://leagueoflegends.fandom.com/wiki/Teamfight_Tactics

PekiCodexJul 1, 20191 votes

I know they are known. I was just stating it as an example, since at LVL 6 it is possible to receive champions of any cost. And I have used the # of champions per cost in order to calculate the champions each player would need to start skewing the %s.

PenPenJul 1, 20192 votes

I already believe those % will skew. For example I had a game where I was getting a lot of 4-drops at level 5 and 6 (around 5 or 6) When quickly scanning the other players I discovered they had a lot of upgraded 1-cost and 2-cost champions AND that they levelled up earlier while I did not. So that meant that they had a small chance of getting 1-drops as option and more % chance of getting 2 and 3 cost as option. Where I had a big chance of getting 1-drops BUT since there weren't much left since everybody already took them I got a lot of 4-drops options instead (since I had a small chance of getting 2, 3 and 4-costs but still there had to be 5 champions available for everyone forcing me to get the 4-drops easily). So although I don't really have proof for this I doubt this was just insane luck.

PekiCodexJul 1, 20191 votes

It is very unlikely that something like that can happen in normal play. Even if someone had five 3-star C1 champions (which is already unlikely), that's only 45 C1 champions, which is not even close to draining the cost pool below 40.

Even at LVL 6 it's possible, even though insanely improbable, to receive five C5 champions in your store. That's 0.5% chance each mind you, but the possibility is there.

What you described could be considered mere luck.

PenPenJul 1, 20192 votes

It isn't unlikely at all what happened. In every game there is a fixed number on how many champions there are for each cost. T1: 39, T2: 26, T3:21, T4:13, T5:10 This is also the reason why you only get 4 options early and later 5. But in the game I played there were insane amounts of nidalee's, Kha'zixs and Vayne's. And everybody was hoarding them but wasn't able to get 3-stars so like 3 or 4 players ended up with 8 versions of that champ. So now there weren't many 1-costs left in the pool and strangely enough those players started leveling up afterwards (yes they actually played very bad). So now with everybody at level 6 while i was still level 5, they had a smaller chance if getting 1-cost as option while I still had a big chance. But like I said since there weren't many left they probably got a lot of 2-cost and 3-cost as option while I still had to get 5 options available it did actually forced me to get a lot of 1-cost options (which i then hoarded up) and afterwards was forced to give me a lot of 4 drop options since it simply couldn't do much else to keep 5 options available for everyone. Keep in mind that normally this would barely be useful since it will only happen in very rare cases like these. But I still hope that somewhere there is a group of 8 people willing to test it since it can very easily be tested with 8 premades. Every player must try to get as many different 1-costs as possible and at level 5, 7 players need to put gold in leveling up without selling any champions and 1-player does nothing. Then 7 players will be level 6 and 1 player will be level 4 or 5. Then those 7 players should hoard up on 2-cost and 3-cost champions. This will then result in that the player that didnt level up will get a lot of 4-cost options.

So, although this isn't very useful in most games you can actually use the drop rates in your advantage. But then again I might be wrong here and this just simply has to be tested.... Hope there is a group of 8 people somewhere reading this and willing to give it a try.

PekiCodexJul 1, 20191 votes

Yes, there is a fixed amount of champions in the pool for each cost... individually. That means there is 39 Vaynes, 39 Garens, 39 Fioras and such.

For comparison, there are:

11 Cost 1 Champions (429 total) 12 C2 Champions (312 total) 12 C3 Champions (252 total) 9 C4 Champions (117 total) 6 C5 Champions (60 total)

So in normal play, there is almost no chance one of the pools being drained. Cost 5 Champions are susceptible to this, but they're rare to obtain already.

Which leaves us with 2 scenarios:

1) The % to obtain a certain cost changes the lower the pool gets, which means the entire level table is just full of shit. 2) You just got lucky getting several cost 4 champions.

I'll let you decide which is more likely to be true.

PenPenJul 1, 20191 votes

Wait how are you guys so certain that it's 39 each T1 champ? I find that very very hard to believe since I can never ever find a certain champion when hyper-rolling and other players already have a few of that champ (it's only a few if there are indeed 429 T1 total) and I ALWAYS find that champion when hyper-rolling after another player sold that champion.

PekiCodexJul 1, 20191 votes

That's because T1 is a special case when it comes to levels. Up to LVL 6 you have the highest chances of rolling them (100-40%), but starting from LVL 6 the odds of drawing them falls significantly (29-10%).

Also despite how many of them there seems to be, there is only a mere 9.09% of drawing a particular champion from the pool. This changes as champions are taken of course, but not significant enough to make a great change, maybe up to 15% at best.

PenPenJul 1, 20191 votes

Please see my other comment above since I am now very certain it isn't 39 of each. See my other comment on Saezio. Would be pretty strange to get 0 or 1 1-cost champions 4 rolls on level 4 (lot of 3-drops of course since the other players had low chance of getting those but still had to get enough options so they had lots of 2-drops). And at level 5 the same thing happened with 4-costs. At level 5 I already had a 2-star katarina, 2-star shyvana and 2 Sol's (now big mister math % genius can you explain that "luck" over and over?) But then I died of course since the other players actually had good comps with of course a lot of 2-star 2-costs since they got those a lot. So like I said a couple of times already: THIS NEEDS TO BE TESTED MORE. But I am alone while trying this and checking the board of 7-players while also hyper-rolling is a bit too much to do in 5-seconds.

SaezioJul 1, 20191 votes

In every game there is a fixed number on how many champions there are for each cost. T1: 39, T2: 26, T3:21, T4:13, T5:10

Just wanted to clarify, that's 39 of each T1 champion not 39 in total. So, 39 nida,39 kha,39 vayne etc.

PenPenJul 1, 20191 votes

Can you show me a link to me where it indeed states 39 each? Since then it should be very easy to get that T1 upgraded to 3-stars. But that would make my whole post indeed kinda stupid but on the other side then we still don't know what happens with champions after a player dies.

SaezioJul 1, 20191 votes

That's what I get from this

PenPenJul 1, 20191 votes

Yeah but that never states each champion. per champion T1: 39.... That could very easily still mean 39 total. But probably is each but that can be easily seen if you count the 1-cost champions of every player in between rounds. Which I will do now in my next game since nobody seems to actually test things and only copy information without actually verifying it.

SaezioJul 1, 20191 votes

"per champion" means for each champion. or it would say per tier.

If it was 39 tier 1 champs total in a game, you could never see more than 4 level 3 tier 1 champs. And I have in fact faced 3 level 3 tier 1 champs while having 2 level 3 tier 1 champs myself. So.

PenPenJul 1, 20191 votes

Was this in the end of the game were other players already died so that their champions came back in the pool probably? And are you sure it was 5 sets of 1-costs and not 4? Since I very very much doubt that. Also see my other comment here as reply to PekiCodex about what happened in my last game when I hyper-rolled very early.

PenPenJul 1, 20191 votes

So I just tested this and I am now very very much convinced it is NOT each. I stacked up only 1-cost minions and went rolling on level 4 and 5 only picking 1-costs. and well well what happened there... I basically only got 2, 3 and 4-cost champions. The only 1-cost champions that came by were the ones I couldn't buy since my bench was full (also swapping those with 1 on my bench and well well there it showed up in the pool while the other didn't anymore). And the next round other 1-drops actually showed up but those were actually the ones other players sold. So I'm actually kinda pissed now since this was VERY fucking easy to find out but nobody seems to want to believe it and are just pointing to a graph stating T1 39 means 39 each.

Oh and have fun abusing this with a premade. 1 person only buying 1-costs (spending all his gold to do so and lose very fast while the other focuses on 2-costs and 3-costs since those will show up more and other players will still try to upgrade the 1-costs but you have to do this at the same time since other players will sell their 1-costs or when someone dies they get back into the pool.

SaezioJul 1, 20191 votes

So, you actually believe there can never be more than 13 level 2 tier 1 champs among 8 people?

PenPenJul 1, 20191 votes

Seems like it. But hard to test alone since other players will sell champions which then other players will buy and therefore will then sell their 1 costs etc. So I will say this 1 more time now: this needs to be tested with a group of 8 premades.... But I very much think it's 39 total and not each.

SaezioJul 1, 20191 votes

I tested :P replied to you on another comment. It is definitely not 39 total.

SaezioJul 1, 20191 votes

There is more than 39 tier 1 champs total in the top 2 spots, so...

PenPenJul 1, 20191 votes

Ok that seems to be true indeed but what about this: I played a game again where I went for a decent early team then selling all my 1-cost champions and buying only 2-cost champions. Again at level 4 I almost only got 2-cost and 3-cost options and at level 5 almost only 1-cost and 3-cost options (at level 5 the other players started selling their 1-cost champions and since I already had a lot of 2 star 2 cost champions there weren't many 2-cost left). Why is this happening so extremely consistent to me then? This can't be just luck if it happened 3 times in a row now.

P.S.
This isn't very good to do since it will fuck up your economy very much but I still ended 3rd this game since I could get all glacials very easily.

PenPenJul 1, 20191 votes

I played again and I still think it isn't 39 each. I went for a semi decent early team which got me on a losing streak but without losing much health (losing all rounds before the first carousel is actually quite good). Then I focused on getting good T2 champions at level 4. And then at level 5 it happened: other players started selling their T1 champions. So I started re-rolling and again i got a lot of T1 and T3 options and not T2. Actually also a decent amount of T4 as well but although it was very tempting to buy them I ignored those to finish my T1 and was picking up some T3 on the way (since I already had a T2 zed and lots of khazix I knew I was going for an assasin comp). I also noticed that other players weren't getting any assasins later on and that I kept a sol somewhere on the way since I had some spare gold and when I later sold it (no pun intended) the next round another player got 2-star sol. Again hard to believe that was yet again pure coincidence.

So now we know that it isn't 39 T1 total but probably not 39 of each T1 either.

SaezioJul 1, 20191 votes

I am guessing all the numbers in your 1st paragraph are random right?

Cause how can those percentages be static when they are changed every time a champion is bought/sold or a player is eliminated

PekiCodexJul 1, 20191 votes

That number is static. It represents odds of getting a champion of that cost, not each champion individually.

Once a cost is selected, a champion of that cost is randomized. The less of a champion in that cost pool, the lower the chances of getting that particular champion.

Of course, that's on LVL 6. The %s change based on level.

SaezioJul 1, 20192 votes

My bad, my brain farted there lol. xD

So at level 6 there is a 2,48% chance to get a Tier 5 champ. Would have thought this was way lower.

Edit: 2,48% per reroll/round

PekiCodexJul 1, 20191 votes

Well, uh, it is. 0.5% at LVL 6. You can see the % table here.

SaezioJul 1, 20193 votes

It's 0,5% per slot though. And we got 5 slots, so its 100%-(99,5%^5)

PekiCodexJul 1, 20191 votes

Oooh, yes, that's correct. [sg-zephyr]

Pierce The VealJul 1, 20191 votes

Except that is not how probability works, and your math is incorrect.

Aside from the probability to get a champion being dependent on what has been taken, that is incorrect. 99.5% is the chance to not roll a tier 5, hence 0.995^5=0.975: you have a 97.5% chance at level 6 to not roll any tier 5 champions. Hence you have a 2.5% chance of rolling 1 or more tier 5 champions.

SaezioJul 1, 20191 votes

I said the exact same thing? Except it isn't 2,5% it's 2,48...%

Edit: I am obviously not accounting for chances dropping because of other people because we don't even know if that happens. (could be that you have a fix 0,5% chance to get a tier 5 and it gives you whatever tier 5 is available, so the chance doesn't drop when others have already gotten some tier 5 champs, what I mean is it is not like a card deck that has specific picks in it, but rather a series of 5 5 sided dices with uneven sides according to each tier's probability)

ThingamajigJul 1, 20191 votes

3 stars are really only important on a few units

Zed Garen Darius WW draven nidalee lulu sej the rest are kinda meh and you should focus more on completing a comp than tunneling on a 3star

PenPenJul 1, 20191 votes

I totally agree with you but a lot of other players don't seem to know this and still try to get 3-stars a lot. Which actually helped me get 4cost-drops pretty easily (see my other comment). So it seems that when everybody is hyper-rolling and leveling up (most of the times after krugs they hyper-roll and the round after they level up to lower the chances of getting 1 costs after they already got what they needed), you actually can sit still like a duck and get higher chances of 4-drops after. Of course this only works if a lot of players do this and you are able to survive (which can be hard since they all have a lot of 3-stars and you aren't buying anything).

ChaosReynJul 1, 20191 votes

I'd even argue getting T3 Lulu doesn't matter. You're overbuffing with T2 Lulu + any mana item as is. I'd argue honestly T3 falls to champions you're hoping to dedicate to carrying you. That means pretty much any wdc, Sol, Nid, Swain, or whichever knight you can turn into the strongest immortal beatstick in game. (Usually, that means Garen or Kayle.)

Also, note that 3 starring a rank 5 is pretty much impossible...so you may as well take them off the list.

ThingamajigJul 1, 20191 votes

3 starring anything above t4 is already asking too much from the RNG gods and 2 star t5s are already bonkers t star kayle is insane in a noble comp

just fought one yesterday i beat him on his 1 star kayle and after he upgraded to 2 stars it was not even close to beating him anymore

3stars are only worth early because they are cheap to get after that it's worth more to level so you can fill in your comp slots id rather have a 4ninja 6 assassin of 2stars than 3ninjas 6 assassiins with 4 assassins as 3stars

KazymandiasJul 12, 20191 votes

{quoted}

3 stars are really only important on a few units

Zed Garen Darius WW draven nidalee lulu sej the rest are kinda meh and you should focus more on completing a comp than tunneling on a 3star

On what do you base that? Is there a set of stats somewhere to show it?

ProfirixJul 1, 20191 votes

Based on the games I've played, when a player is defeated all of their champions are 'retired' and none of the cards make it back into the pot. Otherwise I would have completed an Aurelion Sol 2 after he died; no one else was holding or had one and I rerolled about 25 times at level 8.

PenPenJul 1, 20191 votes

Well since you were at level 8 I assume a lot of other players died already and all of those champions got back in the pool. Meaning that with only 20 % chance of getting a Tier 4 champion per slot and with so many other champions in the pool the chance you will find it is indeed very very low (especially when that was indeed the only other Sol that was in the pool).