Not sure I get the reasoning behind the increased number of runners you are dutching. If you don't think 4 horses can win (I assume you are punting right? as in doing form and such?) then there is no reason to dutch 4 horses.
But, what I will comment on, is this. Your increased unit stake, is out of proportion. Your initial 2.5 figure, increased up to 8.5, spread across 2 runners. Now that 8.5 figure, supposedly is supposed to get back your 2.5 unit loss on the first race, and then make some profit. Now you have 2 horses running for you now, but you are still backing a losing horse in that race. So 1 winner in 2 races, is supposed to pay for 3 bets.
You have increased your level, by 3.4 times, on the first level.
When you get to your 3rd leg, you have 11 units to make up. Yet, you are only betting 25 units. This is an increase of just under 3 times. Less than the original increase, and you are backing 2 more losers. Now, you need 1 winner, to account for 5 losers.
Your next level, is 2.56 times your previous. Yet in real terms, you are trying to make up 36 units. So you have basically gone from 36 units, to 64 units, or less than double, even though your initial increase was nearly 3.5 times.
Now, depending on which horses you are backing, you may back 80% of the market with 4 runners, leaving you with a potential profit of only 20%. 20% of your 64 stake, does not equate to the 36 units you have already lost.
When I ran my staking system, I always increased it by the same level. And it certainly wasn't anywhere near 3 times. I think I had the stake increasing by around 20%. And my stake, was my return to level, not an actual bet stake.
eg, if my stake was 100, it was 100/odds. My goal, was to back whatever, and have the bet represented by the chance of it winning. I did not wish to rely on a level stake on different level of horses.
I think from what you are suggesting, is that your stake is the same regardless of the odds. Meaning, you place more emphasis on higher odds selections. I do not understand how that can work.
My system was about backing the favourite in every race, to a level dependant on their chances of winning. I think I saw 23 losing favourites in a row, and I can guarantee, that no system can handle a run of outs like that, when the goal is to be in profit on the first winner. It is simply impossible.
I don't have a copy of the system I created, but I will give the basics out, as maybe it will invoke some thought.
Each favourite, was recorded in terms of its chances of winning (based purely on the odds I bet it at, right or wrong).
I kept a running total of these odds, as it gave me an indication of how many winners "should" have happened over the duration. I recorded in the column next to it, a running tally of the total winners so far.
Next to that, was an % of expected winners, to actual winners. This was always based on the previous 100 bets, to get a decent average, and to lessen the impact of the increase in stakes.
For example, if over the previous 100 races, you averaged 5.00, you would have an "expected winners" tally of 25. If you had 50 winners, your multiplier would be 25/50, or 50%. Your running stake, would be multiplied by this number. If, on the other hand, you had 10 winners in that time, your multiplier would be 250%.
I also had another multiplier column, to increase the stakes on a loss, and decrease on a win. I had that permanantly set. I think the increase was about 1.2, but can't recall the decrease. Think it was about .7, but that is a guess. I know I had it decrease much quicker. This was the main multiplier, and I would use that running stake, multiplied by the original "expected winners multiplier". So if the stake was up to 250, and you had 50/25 winners, you would bet 50% of the 250 stake.
I had a minimum stake set, and also a maximum stake. The "actual" stake was kept running, so even after a really bad run, it may not neccessarily be yet above the minimum, so you would be still betting to the minimum level, due to the fact that it followed a ridiculously good run, obviously.
The points I am trying to make here:
That is the safest staking system I have ever seen, yet I still had to set maximum levels for it so it didn't get too out of control.
It was designed to increase returns, not to get in profit on a single winner.
It is based solely on expected winners, and betting to the market share of the selection. Not simply a monetary unit regardless of the odds.
When you have a really bad run, your stake isn't increasing at too great a rate.
When I hit those 23 losers in a row, it meant not alot in the scheme of the system. I didn't need to get it all back in 1 go, I had many races to get it back. The time it gets dangerous, is on a substantially bad run, like weeks. And believe me, it happens. There were times I would go 3 months and be down over that whole time.
If a staking plan like this, can get so dangerous, then any plan trying to get back your losses with 1 winner, is a road to ruins.
My suggestion to you, would be to change your stakes, to represent the odds of the horse. You should be betting a higher proportion of that stake, on a horse that has a better chance to win. Makes perfect sense to me?
Then, rather than thinking of a race as either backing 1 horse, or 2 horses, it would be represented by market share instead.
You can back 1 horse at 2.00. I would say you have backed 50% of the market. You could back 2 horses at 5.00, and I would say you have backed 40% of the market.
In bad times, when you have lost alot of units, you don't need the pressure of trying to get that back on a short priced winner, paying you nothing, when you have lost your units backing equal money on them at much worse odds.
Whichever way you go about it, you are going to have massive runs of losses. The higher the market share you take, the bigger the losses will be. Less likely you will lose, but when it happens, you lose big. Makes no difference really, it will happen, to the same proportion. Just exaggerated more the higher market share you take.
Base your increase in stake around expected winners, not monetary losses.
Well, that would be my opinion anyway.
