Mark To Market - Hedging vs Greening Up

[jfdinneen]

In sports trading, seasoned professionals advise newcomers to "bet your edge" in every market. This advice is the layman's equivalent of a mathematician recommending the Kelly Criterion and is, not surprisingly, predicated on the newcomer knowing when he has an edge. Since it is rarely possible to know (for certain) what advantage one has in any specific sporting event, we are, in effect, always "marking to model". There are only two models available to us, though there are many variations - fundamental models (e.g., speed ratings) and technical models (e.g., weight of money). However the initial selection is made, once it is traded (i.e., Buy or Sell) we are in a position to use a "mark to market" approach should the prices move in our favor. You will recall that mark to market is currently in the limelight with respect to financial institutions mishandling of sub-prime mortgage instruments.

An example of how this approach may be used is shown below in the context of an In-Play market on a betting exchange. Assuming that you Buy the selection for $100 at 3.75, you can expect to return $375 if the selection wins. However, what should you do if, during the race, the selection trades at 2.00. Well, if you are marking to market then the market value (see Weighing The Odds in Sports Betting) of your original trade at this new lower price is InitialReturn / (1 - HedgingOdds) = $187.50 giving you an exposure of HedgingMarketValue - InitialMarketValue = $87.50. Depending on your level of risk aversion, you could decide to Sell the selection for $87.50 at 2.00 guaranteeing a return regardless of outcome. This form of hedging is an alternative to "greening up", which effectively closes the open trade at a particular price (i.e., no upside if selection wins). Though I have not tested this approach with Pari-Mutuel betting, it does appear (at a conceptual level) to apply to multi-race exotics.

Apologies, in advance, if this topic has already been covered on this forum or if I have made any errors in calculation.

John

[Robert Fischer]

Always green up.
If you are playing the exchanges you should be making the initial wager with the goal of setting up a green-up opportunity.

[Robert Fischer]

In sports trading, seasoned professionals advise newcomers to "bet your edge" in every market. This advice is the layman's equivalent of a mathematician recommending the Kelly Criterion and is, not surprisingly, predicated on the newcomer knowing when he has an edge. Since it is rarely possible to know (for certain) what advantage one has in any specific sporting event, we are, in effect, always "marking to model". There are only two models available to us, though there are many variations - fundamental models (e.g., speed ratings) and technical models (e.g., weight of money). However the initial selection is made, once it is traded (i.e., Buy or Sell) we are in a position to use a "mark to market" approach should the prices move in our favor. You will recall that mark to market is currently in the limelight with respect to financial institutions mishandling of sub-prime mortgage instruments.

An example of how this approach may be used is shown below in the context of an In-Play market on a betting exchange. Assuming that you Buy the selection for $100 at 3.75, you can expect to return $375 if the selection wins. However, what should you do if, during the race, the selection trades at 2.00. Well, if you are marking to market then the market value (see Weighing The Odds in Sports Betting) of your original trade at this new lower price is InitialReturn / (1 - HedgingOdds) = $187.50 giving you an exposure of HedgingMarketValue - InitialMarketValue = $87.50. Depending on your level of risk aversion, you could decide to Sell the selection for $87.50 at 2.00 guaranteeing a return regardless of outcome. This form of hedging is an alternative to "greening up", which effectively closes the open trade at a particular price (i.e., no upside if selection wins).

Apologies, in advance, if this topic has already been covered on this forum or if I have made any errors in calculation. Though I have not tested this approach with Pari-Mutuel betting, it does appear (at a conceptual level) to apply to multi-race exotics.
John

This is a very interesting topic.
At first glance it was a little hard to see with the font.



with the parimutuel multi-race exotics --- it isn't as clear cut as an exchange.
Again it helps to "set-up" the hedge with your wager.
One way I have played around with is using the top 3 (or less) FAVORITES in the final leg.
Typically you would have a vulnerable fav in one of the first 2(pk3) or 3(pick4) legs - because who wants to bet a pick 3 with all favs??
If alive to the final leg you have a decent expected payoff. If you can handicap down to 2 other possible winners then you can hedge with them. It is hard to cover ALL remaining selections, so it is usually more hedging than greening.

With greening up in an exchange , i find a terrible bet against that i would feel comfortable laying regardless. Then as soon as the market opens I look to lay. He will rise in price before the market closes, often substantially allowing the green-up. Very very rarely you get caught holding the bag, so it pays to be happy with the lay regardless.

This works especially well with NFL plays and season gimmicks as you aren't competing with as many experts at the lay-back game, but it takes some aptitude for handicapping football.

[robert99]

In a financial market the weight of money and mob mentality can overwhelm the fundamentals. Great companies can sometimes be bought for a song and bad companies cost a fortune. For sports, the fundamentals of the sport winner stays the same, as honest money bet has no influence on the result.

For your example, the strategy could be based more on how true you think the prices 3 and 2 are. One is 33% to win and the other 50% to lose. An equal green book does not need any decision on that percentage. A medium level profit is made and move onto the next event. (There may be hundreds of such bets per hour). If you understand the fundamentals better and have the data to prove that then there could be a point in deciding whether the right win percentage is nearer 50 or 33 and adjusting bets to suit. It is not too great a point as you cannot guarantee actually getting any precise price nor at the precise money you need. Will the 2 go lower or will you miss out and it goes up - both provides an additional probability to consider? So a bird in the hand becomes worth more than two in the bush. With PMU, accurate price and profit estimation are additional hazards.

[jfdinneen]

Robert,

I am uncomfortable with the idea of an all-or-nothing trade from a large winning position as it is psychologically invalid (see Prospect Theory). Both greening up (i.e., equalizing position irregardless of outcome) or reducing exposure to zero (i.e., free bet or free lay) are certainly viable alternatives but both options are too costly in the long-run!

You are correct that selection of the hedging price is arbitrary and I accept that, for some traders, this will follow some "mark to model" approach. Alternatively, pre-selecting a "comfort zone" hedging price is easier for most punters.

Marking to market answers the following questions:

what is market value of trade (Buy or Sell) at current price, and
how much exposure should be hedged at current price?
Leaving yourself with profit irregardless of outcome but with a large upside if the selection wins is psychologically very rewarding over the long-term. Ideally, we should always chose strategies that are both mathematically and psychologically valid.

John

NOTE: In original post, InitialReturn / (1 - HedgingOdds) should read InitialReturn / HedgingOdds (transcription error from Excel).

[jfdinneen]

Robert (Fischer),

As you correctly point out, with many sports we are forced to use quasi-hedging (i.e., small probability lose-lose positions not covered).

For example, in golf, you can hedge a Winner market selection (i.e., Buy) with a Head-2-Head market position (i.e., Sell) on the final day but you cannot completely cover the risk of losing both bets. Equally, with multi-race exotics, depending on the prices of your live selections in the final leg you may be forced to cover only some of your exposure. It sounds like you are having some success with the latter.

John

[classhandicapper]

If I have a very strong opinion about the values, then I will rarely make a bet that I believe will lower my very long term return just for some psychological comfort unless the amount is very large. If the amount is very large, sacrificing a few dollars over the long haul isn't quite as bad to me as throwing away tickets. I usually only tend to hedge in situations where my convinction about the values is somewhat fuzzy.

[robert99]

Robert,

I am uncomfortable with the idea of an all-or-nothing trade from a large winning position as it is psychologically invalid (see Prospect Theory). Both greening up (i.e., equalizing position irregardless of outcome) or reducing exposure to zero (i.e., free bet or free lay) are certainly viable alternatives but both options are too costly in the long-run!

You are correct that selection of the hedging price is arbitrary and I accept that, for some traders, this will follow some "mark to model" approach. Alternatively, pre-selecting a "comfort zone" hedging price is easier for most punters.

Marking to market answers the following questions:

what is market value of trade (Buy or Sell) at current price, and
how much exposure should be hedged at current price?
Leaving yourself with profit irregardless of outcome but with a large upside if the selection wins is psychologically very rewarding over the long-term. Ideally, we should always chose strategies that are both mathematically and psychologically valid.

John

NOTE: In original post, InitialReturn / (1 - HedgingOdds) should read InitialReturn / HedgingOdds (transcription error from Excel).

John,

Presume your reply was to my post.

Prospect theory does not apply or is impossible to apply to trading sports.
I gave a simple example of the nearest you might get in estimating where you might be between 33% and 50%. The first one is a probability estimate, the second is a transient market price that may no longer exist. In practice you might make a hundred bets back/lay on one horse and more likely other horses in the same race (as only one winner) and then do the same with in-running bets - all with prices and offers changing within seconds. That way you are better able to get small amounts on at the better prices. Getting the money on right is the practical issue that makes or breaks things. The computer program based on the strategy you apply makes the decisions and makes the bets so there is no psychological angle that effects decisions. Any event in sports that may change the odds is preplanned for in any case.

Greening up is not costly in the long run. It is far less volatile so you can risk far higher sums on each event. Even getting to a green book stage can be an achievement in itself. There is a huge number of events to green up on so if you want higher total returns you bet on more events where greening up is more likely. It would be nice to maximise return per bet by mathematics but you are betting against opponents unknown, arriving at different times, at different points in the queue, cancelling and remaking offers and they are all trying to defeat you rather than take any pre-determined static position. You can offer -they can offer, but how much and at what price you actually achieve is not controllable even at the stage when their prices appear on the screen.

[jfdinneen]

Prospect theory does not apply or is impossible to apply to trading sports

Robert,

Prospect Theory applies to one's own actions whenever you have to make decisions that include uncertain outcomes. As you know, at its simplest, prospect theory states that losses carry a greater psychological weighting than mathematically equivalent wins. Therefore, it does apply to all markets in which you trade - both financial and sports.

However, I agree with you that all greening and/or hedging trades should be pre-planned.

John

[1st time lasix]

Not sure i understand all of the concepts highlighted in this thread....but i tend to 'hedge' pick six tickets that are still alive when a higher odds-"price horse" has already come in for me. Might be with the Pick 4.... pick three or late double with 'marginal' horses I may have left out of the sequence for economy {bankroll} considerations on my ticket in the Pick six..but if i miss in one leg with one I threw out late in the decision process i want to get my original outlay back at least. Some do not agree with the hedging concept....in some cases where a large payout is possible....I do. I use a bet small to win big exotic strategy. Gladly lose the hedge plays if i am set up to win a nice score. I will sometimes hedge a pick four in the same manner if and only if i have already landed a winner at long enough odds to knock most chalk players out of the sequence. It seems logical and smart to me but I have only about four yrs of experience in these multi race horizontal exotics. Still learning......:ThmbUp:

[robert99]

Robert,

Prospect Theory applies to one's own actions whenever you have to make decisions that include uncertain outcomes. As you know, at its simplest, prospect theory states that losses carry a greater psychological weighting than mathematically equivalent wins. Therefore, it does apply to all markets in which you trade - both financial and sports.

However, I agree with you that all greening and/or hedging trades should be pre-planned.

John

John,

I tried to explain with some careful reasoning why it does not apply in practise. Cumulative Prospect Theory might apply (in theory) but again how could you ever decide in practise what cumulative wealth you would ever want to end up with. I make my living trading horseracing and after so long any loss aversion in a single race to me is absolutely zero - so any practised trader is not even on the S value curve.

Psychology certainly comes into making big life changing decisions with uncertain outcomes and PT has explanatory value there. Trading on sports with two teams only has 3 outcomes - zero high risk surprises.

The only uncertainty, once you can estimate "true" prices reliably, as I have said is whether you can get money on at the right amount and price. If I can, I make a small profit each event whatever the result, if sometimes I can't, I make a very small loss or a slightly larger profit. The same process repeats endlessly and is quite tedious and unemotional.
If I (non green screen) hedge, I may make a little bit more or lose a lot more on each event which possibly comes to the same cumulative wealth in the end.

[jfdinneen]

...In practice you might make a hundred bets back/lay on one horse and more likely other horses in the same race (as only one winner) and then do the same with in-running bets - all with prices and offers changing within seconds...

Robert,

Reading between the lines, it would appear that you make your living from scalping horse racing and soccer markets using a betting exchange. If this inference is correct then, firstly, I congratulate you on your success and, secondly, I assert that because you are effectively reducing the time dimension to zero marking to market may be of limited value to you personally.

That said, you will agree that merely because you have discounted the approach in this specific circumstance does not invalidate it in general. Given large open positions, one or more uncertain outcomes, and a minimal time dimension, marking to market hedging is both a mathematically and psychologically valid approach for most traders.

John

[classhandicapper]

I play horses on the exchanges and hedge a lot when I can go "against" a horse at the exchange and bet it at the track at a higher price to lock in a profit either way. There are way fewer opportunities to do that these days, but at one time I was getting several opportunities per day. It was basically free money.

If I understand what you are asking, you want to know whether it makes sense to take both sides if one side or the other has a negative expectation or if you aren't sure whether the market price is an underlay or overlay.

For me the answer is that the vast majority of the time I don't have a strong enough opinion about the values to be sure if the horse is an overlay or underlay at either price. So using the market prices to lock in a profit almost always makes perfect sense.

On the occasions I do have a strong opinion, I tend to stay on one side of the trade unless the amounts are large. However, it is so extremely rare for me to believe I can get good value on both sides, it's practically not even a real world situation. I don't even think I've seen it in horseracing. I can't get specific enough about the values to be sure. I would need a huge odds range. That doesn't happen in horse racing, even at the exchanges (at least for me).

[jfdinneen]

classhandicapper,

Sorry for any confusion. As it happens, I was not asking a question. Instead, I was providing an answer to the hypothetical question as to what size stake to use should you choose to hedge an open position.

It is good to see that you are already arbitraging your plays by taking advantage of the price differences between specific pari-mutuel and exchange markets.

John

[classhandicapper]

classhandicapper,

Sorry for any confusion. As it happens, I was not asking a question. Instead, I was providing an answer to the hypothetical question as to what size stake to use should you choose to hedge an open position.

It is good to see that you are already arbitraging your plays by taking advantage of the price differences between specific pari-mutuel and exchange markets.

John

One of the issues that comes up in pari-mutuel vs. exchange betting is that the exchange odds are locked in, but the pari-mutuel odds are not. There are often significant odds changes on the last flash in the pari-mutuel pools. So typically, you have to contruct the bet so that if you bet "against" in the exchange and the horse loses, you win a very small amount and if the horse wins, you win a lot in the parimutel pool. That way, if the horse's odds fall in the last flash and it wins you will typically at least break even. I have found that he general tendency is for the pari-mutuel odds to fall except on longshots (possibly because other people are engaging in the same activity and placing their bets as late as posible).

[toetoe]

Marvelous thread, boys and girls. :ThmbUp:

[jfdinneen]

toetoe,

If you would like to experiment for yourself with a "Buy Trade & Sell Hedge" position, you can calculate the required hedging stake using the following simplified formula:

HedgingStake = ((InitialStake * (InitialOdds - HedgingOdds)) * (1 - Tax)) / (HedgingOdds * (HedgingOdds - 1))

John

[toetoe]

Thanks. :ThmbUp:

[jfdinneen]

toetoe,

Using market value, the following formulae give the recommended greening odds and stake based on an initial Buy Trade:

GreeningOdds = ((InitialReturn + InitialStake + (InitialStake * InitialOdds) - (InitialStake * InitialOdds * BetfairTax)) / (InitialReturn + (InitialStake * 2) - (InitialStake * BetfairTax)))

GreeningStake = ((InitialStake * (InitialOdds - GreeningOdds )) * (1 - BetfairTax)) / (GreeningOdds * (GreeningOdds - 1))

These calculations assume that you have no other open positions in this market.

John