7

Betting contest based on odd (no money involved)

Comments:
Threaded Linear
#1
Anguibok

Hello guys,
I've been creating an algorithm that give every Tier 1 team an "ELO", based on that ELO I can calculate the chance of every team to win, and then betting odd. The algorithm still need few adjustments to be efficient and probably have a lot of flaws, and the best way to find them is to test it ! And the best way to test it is to make it compete against humain and see if it get destroyed or not.
So I want to organize a little contest (With no money involved of course, i'm not a betting site that will try to take your money), before every match, I will give the odd for every game and you can bet in commentary. All of you receive 1000 point, you can bet 100 point maximum on each game.

So here are the odd for day 1 :
Day one Odd in [], chace of win in ()

  • [1.7] G2 - T1 [2.42] : (G2 59%)
  • [1.61] FPX - FUT [2.63] : (FPX 62%)
  • [1.61] GEN - LEV [2.63] : (GEN 62%)
  • [1.48] TH - DRG [3.07] : (TH 67%)

(If that interest people I can give the precise methodology I used for this algo)

#2
Refrainings
0
Frags
+

fpx beating fut?

#3
Anguibok
1
Frags
+

Yes I know, because China didn't play enough competitive games they are a little bit overrated, You can get easy points by betting on FPX I think (If FPX does good, that will probably be a good point for the bot, and it's even worst for EG that are still ranked at top of NA because the inertia of last year)
I think with more time this problem will be solved. My bet will be FPX loosing to FUT, then being favorite in lower round and still loosing again, so probably free 500 point here...

#21
Refrainings
2
Frags
+

sorry for doubting you king

#28
Anguibok
0
Frags
+

I'm not the queen here... Even I doubted my own child...

Sorry for doubting you princess

#25
welikefortniteandvalorant
0
Frags
+

nah bro was cooking

#17
Anguibok
0
Frags
+

I still think FUT gonna win that but lets be honest, on the first 2 games FPX was better @.@

#4
Anguibok
1
Frags
+

Some hints about the bot.

We have not enough data for chinese team, so top chinese team may be a little bit overated, same for EG, they didn't loose enough to loose the advance they took during last year (I think only EG and FPX are overated, but after 2 loose everything will be fair again)

Actual poweranking of Masters team:

PRX 111
EDG 91
FNC 81
FPX 59
100T 57
GEN 54
TH 35
G2 34
FUT 31
LEV 26
T1 13
DRG -7

If FPX win at least one game the bot will not be that failed

#5
Psion
1
Frags
+

ngl this list is cooked
PRX ✅
EDG ❌
FNC ❌
FPX ❌
100T ❌
GEN ❌
TH ❌
G2 ❌
FUT ❌
Lev ❌
T1✅
DRG✅

#6
Anguibok
0
Frags
+

Keep in mind that what is important here is the value not the rank.
Just take out chinese team of the ranking, thats a point I still have to work on.
What do you think ? I suppose you would put NA team higher ? (If thats happen that will be a huge rebalancing of ELO, and Americas mean ELO will be mush higher)

Also, on who you will bet (and how mush ?)

#7
Psion
1
Frags
+

even removing the chinese teams this still only has PRX and T1 right
PRX ✅
FNC ❌🔽
100T ❌ ⬆️
GEN ❌⬆️
TH ❌🔽
G2 ❌🔽
FUT ❌⬆️
Lev ❌⬆️
T1✅

#8
Anguibok
1
Frags
+

And with Value ? (10 point diff is, 54% Win chance, 50point is 70%, 100 point is 85%, 200 is 97% etc)

#10
Psion
1
Frags
+

the values give way too much to PRX and FNC, its just closer than that (outside of T1 and DRG, they both suck). PRX do not have a 75% chance to beating LEV for example, its more 60-40 PRX

#11
Anguibok
0
Frags
+

Hmmm kinda agree on that, FNC is really high because seems FNC 2023 was the best valorant team ever (With only 2 loose during the seaons)
About PRX I don't think they will win the competition, but I think it's the only team i'm kinda sure they will make top 3, I can't say that for any other team.
FPX steamrolled their own region it's the first time another team beside EDG dominate the region, but I don't know if thats because they improve or because every other team beside EDG is now really trash in China, FPX gonna be a crash test, (Even if they are no way they make top 4)
Lev kinda struggled a lot in playoff and kickoff, I think I would also rank them a little bit higher, but not that mush.

With PRX-FNC-EG being that high I think I shally really consider doing a "soft reset" every year, but that would mean FPX potentially being even higher so... I'm not really sure how to solve this

#12
Psion
1
Frags
+

FPX improved, but even so using formulas is dumb without the eye test. if you want to cooperate and i provide numbers for each team to add an eye test to the calculation i'd be glad to

#13
Anguibok
1
Frags
+

PRX 89
EDG 74
FPX 54
FNC 49
100T 48
GEN 44
TH 29
G2 22
LEV 19
FUT 16
T1 9
DRG 1

With a soft reset every acts

How would that work using "eye" ?

#14
Psion
1
Frags
+

excluding china, this is better, but i still think the disparity between PRX and the lower teams is still crazy high
i'd make my own 1-100 rating or whatever for each team and then we can combine them with percentages (50-50, 60-40, etc) to create a new one
such as(no china to make it easier)
PRX 77
100T 70
GenG 64
FNC 58
TH (with miniboo) 58
Lev 45
Fut 39
G2 34
TH (without miniboo) 24
T1 10
DRG 10
this would lead to, using 50-50s
PRX 83
100T 59
GenG 54
FNC 53.5
Lev 32
G2 28
Fut 27.5
TH(without miniboo) 27
T1 9.5
DRG 5.5

#9
Dolphin
1
Frags
+

100 on G2
100 on Fut
100 on TH
100 on Lev

#19
Anguibok
0
Frags
+

+70 for G2
-100 for Fut
total : 9970
The FPX upset was so weird

#20
Dolphin
0
Frags
+

Your algorithm was right all along, china numba one

#29
Anguibok
0
Frags
+

So after round 1 :

  • 400 spended
    +170 for G2
    +148 for TH
    total : 918
#18
Anguibok
0
Frags
+

I'm not sure to see how we can use that for making odd

  • Preview
  • Edit
› check that that your post follows the forum rules and guidelines or get formatting help
Sign up or log in to post a comment