Part III. Special techniques
In the first and second parts of the publication for modeling used a typical AoF players and the game without special techniques. The purpose of this part is to show that even the special techniques discussed below are in practice ineffective and do not make the game profitable.
Bumhunting
Bumhunting is a type of game in which a player plays against weak opponents only, avoiding strong players.Let's consider choosing a table with a suitable composition of players. Logic dictates that in order to win, you need to have a better loser or tighter player than the average player used in the simulation. In Part I, a player with an all-in statistic of 37% was defined as such. In the following, the term "Tight Player" will be used for players with an All-In statistic of less than 37% and "Loose Player" for players with an All-In statistic of more than 37%. Players with statistics less than a certain specified value are marked with a "minus" sign, players with all-in statistics more than this value - with a "plus" sign. Player statistics are recorded clockwise, starting with Hero. The entry [26-; 37; 58+] means that in place 1 there is a player with 26% or less all-in statistics, in place 2 - a player with 37% statistics, in place 3 - a player with 58% or more all-in statistics (Fig. 4).
For calculations, we will use the player HeroEV and the opponents Oppo with fixed statistics, which were described in the first part. The sessions will be one million hands long. The purpose of modeling is to find such a statistics of opponent's all-in bets, at which the winrate of the opponent will be close to zero, i.e. we will look for some equilibrium state, the change of which leads either to profit or to loss. We will also consider different number of players at the table. The results below show this equilibrium state, and the signs "plus" or "minus" indicate in which direction the statistics of the opponents must change in order for him to be in profit.
The four players at the table are::
- [26–;26–;26–] — a completely Tight table;
- [55+; 55+; 55+] — Fully Tight Table;
- [29–; 51+; 51+] — one Tight and two Loose players;
- [29–; 29–; 51+] — two Tight and one Loose player.
Hero took part in 32-36% of hands.
The sequence of players is important. It is necessary that the tie player is necessarily to the left of Hero. For example, the composition [51+; 29-; 29-] will be very unprofitable in contrast to the composition [29-; 29-; 29-; 51+]. It should be noted that the presence at the table of one or more players with the statistics of all-in 32-39% will also lead to losses.
Three players at the table:
- [27–; 27–] — tight table;
- [55+; 55+] — loose table;
- [30–; 50+] — one Tight and one Loose player.
All-in rates for Hero EV are 39-46%. Again: the tie player must be on the left.
Two players at the table:
- [28–] — tight player;
- [67+] — loose player.
Hero EV will be in the all-in in 54-61% of games.
At first glance, the situation does not look hopeless: a profitable table can still be found. In fact, it is not so. Let's take a full table with timed players [26-; 26-; 26-; 26-]. Based on the distribution in the first part, the probability of meeting a player with all-in statistics less than 26% is approximately 10%. The probability of meeting three such players at the same time will be less than 0.1%. For other full tables, the probability of getting the right lineup is even lower. When playing with three or two players, the probability of finding the right lineup increases, but the advantage will be short: a new player will be quickly placed at this table or one of them will leave the table. I would also like to remind you that here we are talking about long-term statistics of players' all-ins, collected for at least several dozens of hands.
Team Play
Team play in poker is a team game between two or more players who conspire in advance to gain an advantage. Here we will consider a type of team play - collusion. Specifically, players exchange information about their pocket cards.
To model the "conspirators" we use Hero EV, but with the difference that cards known from colleagues were excluded for EV calculation. Let's call it Insider. There can be two or three team players at the table. For modeling the opponents, let's take the same player Oppo. The duration of the session is one million hands.
Let's consider two different variants
The first variant - team players know each other's cards and determine the strength of their pocket cards. The player with a weak hand discards his cards, and the player with a stronger card continues to act as usual, i.e. calculates the EVs of his cards and, if the expectation of profit is positive, goes all-in, otherwise passes. Profits and losses of all insiders are summarized and divided by their number. Thus, the average winrate for an insider is obtained. We will compare this figure with the Hero EV win rate calculated without the team's participation. The results of modeling are shown in Table 6.
| Player composition | Winrate without jackpot BB/100 hand | Winrate with jackpot BB/100 hand | All-in percent |
| Hero, Oppo, Oppo, Oppo | –5.4 | –3.5 | 35.71 |
| Insider, Insider, Oppo, Oppo | –5.6 | –2.9 | 25.11 |
| Insider, Oppo, Insider, Oppo | –4.6 | –1.4 | 29.54 |
| Insider, Insider, Insider, Oppo | –6.5 | –4.6 | 18.61 |
The second scenario is that insiders know each other's cards. The expected profit is calculated based on this information, i.e. the known cards are drawn from the deck and then the EV is calculated. Then, if the expected profit is positive, the insider goes all-in, if it is negative, he discards the cards. There may well be a situation at the table where the insiders play among themselves. The results are summarized in Table 7.
| Player composition | Winrate without jackpot BB/100 hand | Winrate with jackpot BB/100 hand | All-in percent |
| Hero, Oppo, Oppo, Oppo | –5.4 | –3.5 | 35.71 |
| Insider, Insider, Oppo, Oppo | –7.1 | –4.4 | 30.97 |
| Insider, Oppo, Insider, Oppo | –6.8 | –3.8 | 31.48 |
| Insider, Insider, Insider, Oppo | –8.4 | –5.4 | 30.98 |
Based on the modeling data, the following conclusions can be drawn:
- In some situations team play is less unprofitable than playing alone;
- In case of knowing partner's cards, even with a positive EV, it is more profitable to discard a weaker hand than to play these cards together;
- On a full table, two insiders are better than three.
And the main conclusion is that being part of a team does not make the game profitable. If we take into account that the losses are multiplied by the number of insiders, it loses its practical meaning.
Security Policy and Repression
There is another factor that affects the profitability of playing on the GG Network, and that is the security policy and the treatment of professional players.
The security policy in GG Network is simple, it prohibits virtually everything. The punishment for violation is always the same - blocking of the account and confiscation of all funds. Here is an excerpt from the rules (paragraph 1.6): "If there are reasonable grounds to believe that the User is engaging in prohibited activities and behaviors, or is using auxiliary tools and services, the Security Service reserves the right to block and conduct a security check of any account at any time, even during gameplay. The security check may result in, inter alia, restricting access to the User's account, cash shop, gameplay (both cash games and tournaments) or bonus rewards". That is, the account can be blocked at any time, for any reason or no reason. Judging by the discussions of this issue on forums, appeals are practically useless. In most cases, the security service simply ignores them, sometimes giving an arbitrary reason for blocking the account and confiscating your funds. However, given the selfish motivation, this is quite logical.
A professional player is someone who makes money from poker, and GG Network's attitude towards such players, if you put aside their wordy language, is, to put it bluntly, negative. In 2020, they even issued a release about it. Here are a few quotes from it: "A bad pro focuses on creating an unfair advantage over his opponents and exploiting the perceived weaknesses of other players", «We understand that the activities of bad pros are not illegal to the letter of the law. However, as a poker operator with the right to protect the game and the ecosystem, we define these activities as cheating and punish them severely within our rights".
It seems to me that the conclusions here are obvious: if a player wins, he immediately moves into the category of professionals and automatically falls under the repression of the poker room.
Conclusions
You will always lose in the long run, no matter how well you know how to play. Jackpot wins improve the situation a bit, but still don't make the game profitable. A winrate of 3-5BB losses per 100 hands would be a very good result.
With a small number of hands there is a chance to be in profit, but it is always less than 50% and you will more often find yourself losing than winning. As for the possibility of multiplying your bankroll, the chances of achieving this in AoF are less than in French Roulette.
Bumhunting can be profitable, but the probability of finding a table with the right mix of players is very small, almost nil. Team play reduces the loss slightly, but still does not make the game profitable.
Security policies and attitudes towards professional players further reduce the likelihood of profit. If you are lucky enough to make a profit, the important question remains, can you take it back? To summarize, there is no financial or other pragmatic reason to play All-in or Fold on GG Network.
One Comment
олично ! Играю в ШОРТДЕК заметил что все делается для олл инов, просто куча олл иннов из за соотношения банк/бай ин. Регуляров просто нет, ждать АД/АА и прочее нет смысла, блайнды все сьедают. Приходится быть таким же Фишом как и все и играть ПУШ при более менее средних картах, а в холдем 6+ это 80% колоды. ЭТО просто АоФ бессмысленный получается…