“Reinforcement Learning” Sport Betting
“Reinforcement Learning” Sport Betting “IF” Bets and Reverses
I mentioned last week, that if your book offers “if/reverses,” you can play those instead of parlays. Some of you may not know how to bet an “if/reverse.” A full explanation and comparison of “if” bets, “if/reverses,” and parlays follows, along with the situations in which each is best..
An “if” bet is exactly what it sounds like. You bet Team A and IF it wins then you place an equal amount on Team B. A parlay with two games going off at different times is a type of “if” bet in which you bet on the first team, and if it wins you bet double on the second team. With a true “if” bet, instead of betting double on the second team, you bet an equal amount on the second team.
You can avoid two calls to the bookmaker and lock in the current line on a later game by telling your bookmaker you want to make an “if” bet. “If” bets can also be made on two games kicking off at the same time. The bookmaker will wait until the first game is over. If the first game wins, he will put an equal amount on the second game even though it has already been played. “Reinforcement Learning” Sport Betting
Although an “if” bet is actually two straight bets at normal vig, you cannot decide later that you no longer want the second bet. Once you make an “if” bet, the second bet cannot be cancelled, even if the second game has not gone off yet. If the first game wins, you will have action on the second game. For that reason, there is less control over an “if” bet than over two straight bets. When the two games you bet overlap in time, however, the only way to bet one only if another wins is by placing an “if” bet. Of course, when two games overlap in time, cancellation of the second game bet is not an issue. It should be noted, that when the two games start at different times, most books will not allow you to fill in the second game later. You must designate both teams when you make the bet.
You can make an “if” bet by saying to the bookmaker, “I want to make an ‘if’ bet,” and then, “Give me Team A IF Team B for $100.” Giving your bookmaker that instruction would be the same as betting $110 to win $100 on Team A, and then, only if Team A wins, betting another $110 to win $100 on Team B.
If the first team in the “if” bet loses, there is no bet on the second team. No matter whether the second team wins of loses, your total loss on the “if” bet would be $110 when you lose on the first team. If the first team wins, however, you would have a bet of $110 to win $100 going on the second team. In that case, if the second team loses, your total loss would be just the $10 of vig on the split of the two teams. If both games win, you would win $100 on Team A and $100 on Team B, for a total win of $200. Thus, the maximum loss on an “if” would be $110, and the maximum win would be $200. This is balanced by the disadvantage of losing the full $110, instead of just $10 of vig, every time the teams split with the first team in the bet losing.
“Reinforcement Learning” Sport Betting
As you can see, it matters a great deal which game you put first in an “if” bet. If you put the loser first in a split, then you lose your full bet. If you split but the loser is the second team in the bet, then you only lose the vig.
Bettors soon discovered that the way to avoid the uncertainty caused by the order of wins and loses is to make two “if” bets putting each team first. Instead of betting $110 on ” Team A if Team B,” you would bet just $55 on ” Team A if Team B.” and then make a second “if” bet reversing the order of the teams for another $55. The second bet would put Team B first and Team A second. This type of double bet, reversing the order of the same two teams, is called an “if/reverse” or sometimes just a “reverse.”
A “reverse” is two separate “if” bets: “Reinforcement Learning” Sport Betting
Team A if Team B for $55 to win $50; and
Team B if Team A for $55 to win $50.
You don’t need to state both bets. You merely tell the clerk you want to bet a “reverse,” the two teams, and the amount.
If both teams win, the result would be the same as if you played a single “if” bet for $100. You win $50 on Team A in the first “if bet, and then $50 on Team B, for a total win of $100. In the second “if” bet, you win $50 on Team B, and then $50 on Team A, for a total win of $100. The two “if” bets together result in a total win of $200 when both teams win.
If both teams lose, the result would also be the same as if you played a single “if” bet for $100. Team A’s loss would cost you $55 in the first “if” combination, and nothing would go onto Team B. In the second combination, Team B’s loss would cost you $55 and nothing would go onto to Team A. You would lose $55 on each of the bets for a total maximum loss of $110 whenever both teams lose. “Reinforcement Learning” Sport Betting
The difference occurs when the teams split. Instead of losing $110 when the first team loses and the second wins, and $10 when the first team wins but the second loses, in the reverse you will lose $60 on a split no matter which team wins and which loses. It works out this way. If Team A loses you will lose $55 on the first combination, and have nothing going on the winning Team B. In the second combination, you will win $50 on Team B, and have action on Team A for a $55 loss, resulting in a net loss on the second combination of $5 vig. The loss of $55 on the first “if” bet and $5 on the second “if” bet gives you a combined loss of $60 on the “reverse.” When Team B loses, you will lose the $5 vig on the first combination and the $55 on the second combination for the same $60 on the split..
We have accomplished this smaller loss of $60 instead of $110 when the first team loses with no decrease in the win when both teams win. In both the single $110 “if” bet and the two reversed “if” bets for $55, the win is $200 when both teams cover the spread. The bookmakers would never put themselves at that sort of disadvantage, however. The gain of $50 whenever Team A loses is fully offset by the extra $50 loss ($60 instead of $10) whenever Team B is the loser. Thus, the “reverse” doesn’t actually save us any money, but it does have the advantage of making the risk more predictable, and avoiding the worry as to which team to put first in the “if” bet.
(What follows is an advanced discussion of betting technique. If charts and explanations give you a headache, skip them and simply write down the rules. I’ll summarize the rules in an easy to copy list in my next article.)
As with parlays, the general rule regarding “if” bets is:
DON’T, if you can win more than 52.5% or more of your games. If you cannot consistently achieve a winning percentage, however, making “if” bets whenever you bet two teams will save you money.
For the winning bettor, the “if” bet adds an element of luck to your betting equation that doesn’t belong there. If two games are worth betting, then they should both be bet. Betting on one should not be made dependent on whether or not you win another. On the other hand, for the bettor who has a negative expectation, the “if” bet will prevent him from betting on the second team whenever the first team loses. By preventing some bets, the “if” bet saves the negative expectation bettor some vig.
The $10 savings for the “if” bettor results from the fact that he is not betting the second game when both lose. Compared to the straight bettor, the “if” bettor has an additional cost of $100 when Team A loses and Team B wins, but he saves $110 when Team A and Team B both lose.
In summary, anything that keeps the loser from betting more games is good. “If” bets reduce the number of games that the loser bets. “Reinforcement Learning” Sport Betting
The rule for the winning bettor is exactly opposite. Anything that keeps the winning bettor from betting more games is bad, and therefore “if” bets will cost the winning handicapper money. When the winning bettor plays fewer games, he has fewer winners. Remember that the next time someone tells you that the way to win is to bet fewer games. A smart winner never wants to bet fewer games. Since “if/reverses” work out exactly the same as “if” bets, they both place the winner at an equal disadvantage.
Exceptions to the Rule – When a Winner Should Bet Parlays and “IF’s”
As with all rules, there are exceptions. “If” bets and parlays should be made by a winner with a positive expectation in only two circumstances::
When there is no other choice and he must bet either an “if/reverse,” a parlay, or a teaser; or
When betting co-dependent propositions.
The only time I can think of that you have no other choice is if you are the best man at your friend’s wedding, you are waiting to walk down the aisle, your laptop looked ridiculous in the pocket of your tux so you left it in the car, you only bet offshore in a deposit account with no credit line, the book has a $50 minimum phone bet, you like two games which overlap in time, you pull out your trusty cell 5 minutes before kickoff and 45 seconds before you must walk to the alter with some beastly bride’s maid in a frilly purple dress on your arm, you try to make two $55 bets and suddenly realize you only have $75 in your account.
As the old philosopher used to say, “Is that what’s troubling you, bucky?” If so, hold your head up high, put a smile on your face, look for the silver lining, and make a $50 “if” bet on your two teams. Of course you could bet a parlay, but as you will see below, the “if/reverse” is a good substitute for the parlay if you are winner.
For the winner, the best method is straight betting. In the case of co-dependent bets, however, as already discussed, there is a huge advantage to betting combinations. With a parlay, the bettor is getting the benefit of increased parlay odds of 13-5 on combined bets that have greater than the normal expectation of winning. Since, by definition, co-dependent bets must always be contained within the same game, they must be made as “if” bets. With a co-dependent bet our advantage comes from the fact that we make the second bet only IF one of the propositions wins.
It would do us no good to straight bet $110 each on the favorite and the underdog and $110 each on the over and the under. We would simply lose the vig no matter how often the favorite and over or the underdog and under combinations won. As we’ve seen, if we play two out of 4 possible results in two parlays of the favorite and over and the underdog and under, we can net a $160 win when one of our combinations comes in. When to choose the parlay or the “reverse” when making co-dependent combinations is discussed below.
Choosing Between “IF” Bets and Parlays “Reinforcement Learning” Sport Betting
Based on a $110 parlay, which we’ll use for the purpose of consistent comparisons, our net parlay win when one of our combinations hits is $176 (the $286 win on the winning parlay minus the $110 loss on the losing parlay). In a $110 “reverse” bet our net win would be $180 every time one of our combinations hits (the $400 win on the winning if/reverse minus the $220 loss on the losing if/reverse).
When a split occurs and the under comes in with the favorite, or over comes in with the underdog, the parlay will lose $110 while the reverse loses $120. Thus, the “reverse” has a $4 advantage on the winning side, and the parlay has a $10 advantage on the losing end. Obviously, again, in a 50-50 situation the parlay would be better.
With co-dependent side and total bets, however, we are not in a 50-50 situation. If the favorite covers the high spread, it is much more likely that the game will go over the comparatively low total, and if the favorite fails to cover the high spread, it is more likely that the game will under the total. As we have already seen, when you have a positive expectation the “if/reverse” is a superior bet to the parlay. The actual probability of a win on our co-dependent side and total bets depends on how close the lines on the side and total are to one another, but the fact that they are co-dependent gives us a positive expectation.
The point at which the “if/reverse” becomes a better bet than the parlay when making our two co-dependent is a 72% win-rate. This is not as outrageous a win-rate as it sounds. When making two combinations, you have two chances to win. You only need to win one out of the two. Each of the combinations has an independent positive expectation. If we assume the chance of either the favorite or the underdog winning is 100% (obviously one or the other must win) then all we need is a 72% probability that when, for example, Boston College -38 ½ scores enough to win by 39 points that the game will go over the total 53 ½ at least 72% of the time as a co-dependent bet. If Ball State scores even one TD, then we are only ½ point away from a win. That a BC cover will result in an over 72% of the time is not an unreasonable assumption under the circumstances.
As compared to a parlay at a 72% win-rate, our two “if/reverse” bets will win an extra $4 seventy-two times, for a total increased win of $4 x 72 = $288. Betting “if/reverses” will cause us to lose an extra $10 the 28 times that the results split for a total increased loss of $280. Obviously, at a win rate of 72% the difference is slight.
Rule: At win percentages below 72% use parlays, and at win-rates of 72% or above use “if/reverses.” Article Source: http://EzineArticles.com/614359