Placepotting

I've been promising a blog on placepots for a while now, and this seems like as good a time as any to produce one.  Why?  Well, because for the first time, I feel that I have a handle on how to produce winning placepots.  I'm certainly winning at the moment and am up over £200 since January.  I haven't had a truly big win yet, but I'm hopeful.  In the mean time, I've picked up some meaningful prizes, most recently £152 to a £24 stake at Great Yarmouth. 

I should warn you, this is by a margin the longest post I've yet written.  So my advice is to save it until you have some time to read it in detail.  I do hope you find it helpful, and please make use of the comments box, email or twitter to let me know your thoughts.


I'm going to talk you through how I think about the placepot.  Beginning with the analysis of whether to even bother going through the card, based on the "shape" of the card, then an analysis of how to construct shortlists, and a bet, before deciding whether to bet at all.  I'll also talk a bit about certain types of races, and how I typically bet on them.  First, though, a brief bibliography, beginning with the best book on gambling I've ever read.

A little bit of guidance:

I was always fascinated by the ideas of pool betting, but I didn't start winning until I went looking for guidance.  There's a lot of rubbish out there.  There is, however, one tremendous book - Exotic Betting: How to Make the Multihorse, Multirace Bets That Win Racing's Biggest Payoffs.  Written by Steve Crist, one of the Daily Racing Form's finest, it has a strong American focus.  However, the principles it espouses work just as effectively here - with the advantages that we don't pay tax, and that the minimum stake on the placepot is far lower than on their "exotics".  The one problem is that, for multirace betting (as opposed to forecast/tricast betting, and the US equivalents) it focuses on win-only betting.  However, the differences are relatively straightforward.  I'd urge you to invest in this book; like all brilliant ideas, the concepts he suggests are so straightforward that it is hard to believe you hadn't thought of them yourself.

On the placepot itself, the best guide I've found is written by Gary, a blogger I've been following for years over at nag-nag-nag.  He starts with the nuts and bolts of how the bet works, and how to place it, which I won't bother to repeat here, as well as providing his own insights into how the bet can be made to pay.

I'd also recommend looking at the totepool live information screen, especially as bets are going on.  This gives an invaluable insight into where other people's money is going - the money you need to take off them.  The week's history, with results as well as placepot payouts, will provide some context for what I call the "shape" of a placepot. 

The "shape" of placepots:
I've used the word "shape" to define a placepot bet a number of times now.  I don't know if this is how other people think about racecards, or if it is a concept that will make sense.  To me, though, there is a certain character to every racecard, and certain characteristics make it easier, or harder, to make a profit from placepot betting.  When I sit down and analyse a potential placepot bet, my first though process is very quick (less than five minutes).  Analysing the shape of a card is, I suspect, a matter of experience; it is certainly something I do quickly and confidently these days, and I do have a fairly good track record.  I was terrible at this when I started. 


Going through the thought exercise of shaping a card is a good discipline even if you don't intend to place a bet.  The more often you do it, the better - you can also combine it with scanning the cards for horses you want to follow, and it motivates you to examine the results, and spot patterns.  So, what do I do?


Big and small meetings:
This is the first consideration.  A big meeting typically means a bigger pool.  Saturday meetings have large pools, particularly the televised meetings.  Festival pools are often huge.  These meetings, too, will often have "dumber" money in the pools; that is, the proportion of the pot provided by punters with limited form knowledge, or small numbers of lines (or both) will be higher, increasing the edge for the thoughtful punter.  However, these meetings are often highly competitive, and winning pools can be tough. 

My preference is to look for "uncompetive" days at big meetings - the rare days when I can find one or two horses that I really fancy against the field in several races, allowing me to back large numbers in other races without the perm getting out of control.  A good example was the Friday of Aintree's Grand National meeting, in which I singled Big Bucks, Wishfull Thinking and Zarkandar in a single perm of 72 lines (1x1x1x3x4x6) for £36.  6 winning lines paid £320+, even with an unusually low (for Aintree) payout of £107.

On the other hand, small meetings typically have smaller, less competitive fields, and smarter competition.  Here, I'm looking to play if I can see several weak favourites, and keep a relatively small perm whilst seeing some other favourites beaten (say, 2x2x3x2x1x1, as at Yarmouth last week, where I had one winning £1 line from a £24 bet, for £152).


The betting order:
Having viewed the card as a whole, we swiftly go through the six races that make up the placepot.

One revelation from Crist's book was that it is the betting order, not the price, that is the chief determining factor in the inclusion of horses in other people's placepot.  Think of two ten horse races.  One has a 1/2 favourite, a 3/1 second favourite, and the rest at 6/1 or higher.  The other has market leaders at 5/2, 3/1 and 7/2.  You'd expect the 1/2 favourite (67% chance of winning, if we ignore the bookies cut) to be around 67% of the pool, and the 5/2 favourite to have about 28%.  Crist's argument, and the live information board supports this, is that the 1/2 favourite will probably be underbet, whilst the 5/2 favourite will probably be overbet.  I'd expect (and I have no statistical evidence for this) to see approximate 55% and 40% of the market, respectively, on those two favourites.

I would, therefore, be happier opposing favourites in a card with lots of longish priced favourites (heritage handicaps and festival races are classic "overbet favourite" races - amazing numbers of punters use the "favourite" button, the last resort of the coward (or the tragic resting place of the backers of non-runners)) but I don't shy away from backing odds-on favourites as bankers.  You certainly need to know what your opinion on every favourite (or likely favourite, remember that betting forecasts are dodgy) is before you decide whether to bet.

Field size:
This is to some extent linked to meeting size, but there are differences.  Some conditions (fast jumping ground, for example) can lead to very small fields. Some races at big meetings can have very small fields (novice chases, for some reason, seem to fall into this category very often).

More importantly, however, it changes the way in which we think about place betting, as opposed to win betting.  Smaller fields means fewer places (4 horses or fewer means that you need to find the winner), and can, counter-intuitively, mean that more punters are knocked out of the pool.   It may be hard to find the winner of a 16 horse handicap, but 2 sensible selections gives you an excellent chance of a placed horse - whereas you need the winner in a 4 horse race.

As I write this, a six race card at Huntingdon has just paid £2,800+ due primarily (I suspect) to favourites finishing unplaced in four- and six-horse races.  Make sure you have a clear opinion on small field races before you take a stance on the card.  Don't expect to be backing fewer horses automatically in smaller horse races - that's what everyone else is doing.

Estimating returns:
It is almost impossible to estimate a return on a placepot.  So much depends on whether specific favourites are placed or not.  However, in thinking through the factors above you can with some reliability form a ballpark estimate of your own likelihood of winning, and the likely dividend.  The two things are related. 

So, for example, you may think defensively about a possible perm: I can't find many bankers, and I need extensive coverage in at least these two races... I can't form a perm with fewer than 108 lines, and that means 20p lines is the most I can afford.  That's fine, though, because if a few short priced horses come in, I'll soon be up to 2x2x2x1x1x1 surviving, so if it is a small pool, I could easily be looking at 1.6 shares or more.  On the other hand, if I only get 1 line through, but I see the odds-on shot beaten in one of the first two races, 0.2 lines in a big pool would be fantastic.

A more aggressive view of a placepot is: There's three horses who can't lose, and I can get plenty of coverage elsewhere.  If I take on the other three favourites, I can have a 1x1x1x2x4x4 perm for £32 at £1 lines, and I only need one favourite beaten for the payout to be decent for one surviving line.

Both of these approaches are valid.  You do need to be correct in assessing your bankers, and you do need to be realistic (and experienced) in assessing likely outcomes.  Finally, of course, there is no guarantee that you'll be right on any given day - but you can be right more often than the market if you plan your perming around the shape of the card and your estimated return.

Should you think about betting?:
By this point, you've formed a crude impression of whether you fancy the key favourites or not (in my case, going on the timeform summary and my own rudimentary form knowledge), how you view the shape of the card, and your likely perm approach/staking size - coupled with a vague expectation of estimated returns under different result outcomes. 

The key decision now is, should I bet?  My own rule of thumb is to only bet if I expect to win 6 times my stake (for example, if I had 48 50p lines (a typical stake, for me, of £24) I would be hoping to win approximately £150 - and I don't care if I do that with one line surviving in a £300 payout, or 12 in a £24 payout).  This can only ever be a crude estimate, but my experience is that I'll get a placepot payout roughly once in five attempts (though it has been 1:3.5 this year); so getting six times my money back is necessary to maintain profit.

I quite often don't bet.  I aim to get better at not betting this season.

The only other point to stress at this stage in the process is that you cannot become wedded to your ideas.  You might entirely change your mind about a favourite through watching a video, coming across a trainer, jockey, or draw statistic, or just through looking again at the opposition.  You might find a non-runner at the very end that ruins your plans.  Every time you rethink your mental estimated return, or change your rough perming plan think again: should I bet?


How to bet:
If you decide to bet, you need to work out, obviously, who you should be betting on.  This is in two parts - picking selections in indvidual races, and putting your bet together.  Crist is truly extraordinary on the importance of putting your bet together, and my summary below does not do him justice.  I think that the actual selection of horses is something that punters are generally quite good at, but I have a few placepot-specific tips that I'll return to.  For now, though, let's focus on putting your perm together.


To bank or not to bank?:
The essence of a placepot is to survive, to still be there at the bitter end.  We've all had horses disappoint, and (especially in jumps racing) that moment when you realise your selection will not place can be sickening.  The temptation, then, is to put at least two horses in every race selection.  There are two reasons not to do this, however, and they are related.
1. There's always another day. So your banker got beaten?  Don't worry about it.  You only need to win once in five attempts to do very nicely on the placepot. 
2. It is mathematically stupid.  Consider the following two perms:
a) 1x1x2x2x3x3 @£1 = £36. 12 horses selected.
b) 2x2x2x2x2x2 @£1 =£64. 12 horses selected.
c) 1x1x2x2x4x4 @£1 = £64. 14 horses selected.

Option b contains no bankers.  It might, on very rare occassions, be impossible to find a banker.  If you're determined to bet regardless, you'd better be sure of high returns for a win, because your perm costs will spiral (or you'll divide your stake down to very low levels).  I avoid bankers, sometimes, at the Cheltenham Festival.  Everywhere else, if I can't find at least one banker, I don't bet.  The reason is shown above; with two bankers, you can either have more horses running for you (option c) or far less cost for the same horses (option a).  This is because the number of horses you select is additive, whereas your perm is multiplicative. 


How to write a perm (or perms):
Writing perms is broadly a question of understanding the maths above, your own staking limits, and the shape of the card.  Although this varies based on individual race selections, I am most comfortable, and most successful, within a few perm types:
Single banker, several doubles: e.g. 1x2x2x2x2x3 (48 lines).  With only a single banker, keeping the number of lines to a sensible number means being fairly restricted in other races.  However, in uncompetitive cards, this is often possible.
Double banker, extra coverage: e.g. 1x1x2x2x3x5 (60 lines). A second banker means that it is possible to gain quite extensive coverage in one horrible race without ending up with a huge card.
Single banker, low stake coverage: e.g. 1x2x2x3x4x6 (288 lines).  In this example, I reduce my stake to allow wider coverage in a couple of particularly trappy races, with the hope that I'll have several lines returned.  This is the method that I will be using at Ayr tomorrow (which I'll blog on if I can).

Something else that I learn from Crist is that it is often wise to write two perms, or more.  I have started to do this, and it is something I will do more of - although it is perhaps more valuable in the context of win-only multirace betting.  The principle is that, if we see two weak short-priced favourites, it is unlikely that both will be beaten - however, we are confident that one will be beaten.  We can therefore construct two perms, each opposing one weak favourite.  This is perhaps easier to explain with an example and it is something I'll return to when I have time to blog an example of this before I place the bet.


Is it worth betting?:
Again, return to the key question: now I've seen the perm I want to bet, and really thought about the card, is it still sensible to place the bet?  Don't do it  unless you really have to. 

Also, wait until the last minute and check for non-runners.  They really are the pits for the placepot bettor.  While you're waiting, check the market moves, and see if it impacts on your opinions in any way.  If you're at the track, have a look at the runners for the first.  Don't be afraid to change your mind.  In my opinion, it is superstition, and only remembering the missed winners, that puts bettors off flexibility.


Certain types of races:
Whilst I can't go through selection in great detail, there are a few key types of races which I approach differently.   What follows is more in the nature of the occassional tip than anything deeply earth-shattering.

Level stakes (non-novice):
I don't have any specific statistics in mind (though they're out there) but level stakes races run most closely to form, and have the best records for favourites (remember, we're concerned with strike-rate (or place rate) not level-stakes profit, as it is the betting order not the price that affects the bet).  Once horses have reasonably exposed form and are running off level weights, you should be looking to find your bankers here wherever possible. 

One of the most confusing conversations I've ever had with a normally smart racing friend (let's name names: with Jack) was when he asked me, after the race, why I'd singled Frankel in the 2,000 Guineas.  Well, because of what he just did, and because there was no other possible banker on the card.  I did the same with Big Bucks in the World Hurdle.  There's a sense that big races are more competitive because they contain household names, and because we think about the chances of the outsiders.  They aren't more competitive.


Novices and juveniles:
The key here is reliability.  Jumps racing, in particular, is a nightmare for the placepot - one stumble from a banker and you're out.  So be very careful putting a horse forward as a banker unless you really know it can jump (and if it is a novice chaser, that means jumping fences; hurdle technique counts for nothing).

On the flat, I have no problem singling horses who have run before in juvenile/3yo maiden races, but I wouldn't single a debutant.  So much can go wrong, and there are lots of talking horses in the flat game.  Remember, too, that experience counts for a great deal - places are often "nicked" by experienced horses finishing ahead of first-timers who are "staying on after the penny dropped" at the end.  This is a major difference from jackpot selections, where I'd like to have almost all debutants in my list, together with the best form horses.


Handicaps:
In my opinion, handicaps are the place to be opposing favourites, simply because they are typically the races which have the greatest competition, and thus the smallest differences between the chances of horses at different positions in the betting order.  You shouldn't oppose favourites on principle, but looking for horses who might unplace them is a good idea.

Remember, too, that the money in placepots is (or appears to be) placed by less sophisticated punters than in win markets.  So think about avoiding popular horses, and taking advantage of subtler indicators of chances.  In particular, it seems to me that going preferences are under-regarded in placepot markets, that the draw is not given enough attention, and that avoiding "name" jockeys and trainers can often lead to a contrary position - without needing to back rank outsiders.

In truly competitive handicaps, it is enough to simply have one line survive.  Don't be a hero - four selections on four of the front eight in the Lincoln is probably about right. The favourite and two 40/1 shots probably isn't.

----

So that's it.  In essence, that's how I play the placepot.  If time permits, I'll return with smaller posts, looking at how I will be placing my bets on the meetings I like tomorrow (Ayr, and possibly Beverley, I haven't done the detailed work yet).  If not, I'll run a blog at some later date.