Allwin Skill

[Bent Copper]

By some strange coincidence, the Royal Institution Christmas Lecture yesterday was all about Chaos theory. It seems Guest was right, and the lecturer had bouncing balls, magnetic pendulums and double pendulums in the studio to demonstrate the effect of a very small change in starting position leading to a completely unpredictable result. It was most enlightening, especially in view of this current discussion.

One thing we hadn't considered is turbulance around a spinning ball, and even that is significant. Apparently, even the air turbulance itself is chaotic, and there is currently a $1 million prize for anybody who can come up with a formula to predict the behaviour of a spinning ball flying through the air. So what chance have we got of sorting it out? The most we receive for getting it right is 6d!

[pennymachines]

But unless chaos predominates, it may not be as significant as you think. Clearly all these hard to predict interacting forces of ball spin and air turbulence apply equally to football. If that means it's a game of pure chance, the players really are overpaid.

[Bent Copper]

Chaos is much more significant in Allwins than it is with games like football because of the multiple obstacles in the path of the ball and because of the small target areas. Without these obstacles, the trajectory of a football is unpredictable, but still remains within certain limits. But once a ball hits an obstacle, then it will branch off in a completely different direction, so any minor inaccuracy will have a devastating effect. That's where Chaos really rules!

Obviously, the skill/luck ratio is easy to determine in football (or any other game for that matter) just by prescribing the width of the goals, or target. Football is a game of skill, simply because the goals are wide enough apart to take these unpredictable effects into account. Just look how wide the goals are compared to the width of a football! Now look at how wide an Allwin cup is compared to the ball. If football goals were only just wide enough to allow a ball through (as in an Allwin) then it certainly would be a game of pure luck, and would be unplayable. By comparison with football, a snooker pocket is much smaller in relation to the size of a snooker ball, but again the proportions have been arrived at to allow skill to predominate. It is obvious that a sport must be designed to allow skill to predominate, but not so a fairground game.

So I think these comparisons between an Allwin and 'normal' ball sports are irrelevant, because the targets have been designed to allow skill to predominate in the case of sports, whereas the opposite is true in the case of the Allwin. The Allwin manufacturers wanted luck, and not skill, to predominate. They did this by making the target areas as small as was necessary (but not so small as to make the game look impossible) and by putting deflectors in the path of the ball.

So I think this sums it up really. We have already agreed (I think) that multi-cup Allwins are pure luck because of the small target areas. If you don't hit the ball straight into a cup, then after that it's down to pure luck as to whether it will bounce into one of the other cups or not. The disagreement is with Allwins with larger target areas, or wide 'all-winning' galleries. Clearly, it would be quite possible to design an Allwin where skill did predominate. All you would have to do is to remove all the deflecting pins and adjust the size of the target(s) to favour skill rather than luck. But the Allwin manufacturers were not playing that game and such an Allwin would have been a commercial disaster. Nobody can say that no such Allwin exists, and we've heard from some people who say that it does. That's fair enough, there are bound to be exceptions, and it is interesting to hear about them. But I still maintain that the standard Allwin configuration is a game of pure chance.

Happy New Year to all.

[pennymachines]

I'm not certain a multicup is pure luck. I think you can slightly improve the odds by dropping the ball on the "sweet spot" - near to the uppermost centre cup. If you hit it, it's a hole in one; and if you don't, there's a better chance of the ball finding its way into one of the cups below than if it had fallen from one side. This applies particularly to the large 24 Cups.

To make the football analogy more precise, consider a game decided purely on penalty kicks. Like an allwin, the players get a single strike of the ball.

Just look how wide the goals are compared to the width of a football! Now look at how wide an Allwin cup is compared to the ball.

But that's not really comparing like with like. The entire allwin gallery defines the "target", just as the goalposts do in football. This can be quite wide compared to the width of the ball. What's more, the ball and gallery are confined to a single plane, whereas a footballer has to control the ball in three dimensions and may shoot over the crossbar as well as either side of the goalposts. And isn't the deflecting pin (goalkeeper) that tries to get in the way of the ball as much an obstacle as the static allwin gallery pins?

My feeling is that the skill element on some Oliver Whales and Wonders allwins may approach zero, but is much higher on others, as well as those by Bryans (as discussed), Ruffler & Walker (twelve-row column-fill allwins,) Machine Mart, Parkers, Saxony (German-made) and others. On the Brenner Skill Fun the "goalposts" are moved each time you play, continuously varying the skill/luck ratio (and the operator can set the extent to which it does so).

[Bent Copper]

I didn't quite concede that a multicup is pure luck either. I think you can slightly improve the odds by dropping the ball on the "sweet spot" - near to the uppermost centre cup. If you hit it, it's a hole in one; and if you don't, there's a better chance of the ball finding its way into one of the cups below than if it had fallen from one side. This applies particularly to the large 24 Cups.

I think this is just based on wishful thinking, and I would be very surprised if it makes any real difference where the ball lands, or even that it's possible to 'aim' the ball with any degree of accuracy. For instance balls that hit the right-hand cups are more likely to bounce left and vice versa. So balls will tend to bounce towards the other cups wherever they land.

But that's not really comparing like with like. The entire allwin gallery defines the "target", just as the goalposts do in football. This can be quite wide compared to the width of the ball.

Quite wide? Even in the best case, the gallery is only 12 balls wide. I suspect that football goals are considerably wider than 12 footballs. And the entire gallery doesn't define the target anyway because there are only 12 distinct positions that the ball can enter the target without bouncing off. If there are any mathematicians amongst us, they will be able to tell us exactly how much the target area is reduced by those 12 pins. I suspect it is reduced to a very small fraction of the total width.

And isn't the deflecting pin (goalkeeper) that tries to get in the way of the ball as much an obstacle as the static allwin gallery pins?

How can a single moveable pin ever be anything like as effective as 12 static pins precisely set up to be the exact width of the ball? If you wanted to replicate an Allwin on a football pitch, you would need to set the goal posts to the width of only 12 footballs (or even less) and you would need 12 goalkeepers standing side by side with a gap of one football width between them. (Obviously, the goalkeepers are not allowed to move!) Such a game would be impossibly difficult to play, and the only reason that an Allwin is playable at all is because it is a vertical playfield and losing balls may or may not get another chance to hit the target somewhere else, depending on luck.

My feeling is that the skill element on some Oliver Whales and Wonders allwins may approach zero, but is much higher on others, as well as those by Bryans (as discussed), Ruffler & Walker (twelve-row column-fill allwins,) Machine Mart*, Parkers, Saxony (German-made) and others. On the Brenner Skill Fun the "goalposts" are moved each time you play, continuously varying the skill/luck ratio (and the operator can set the extent to which it does so).

That's as may be, but let's confine this discussion to ordinary Allwins rather than 'specials' otherwise we'll just cloud the issue and get nowhere. (We'll get nowhere anyway, but at least let's get there as fast as possible!)

*PS: I didn't know that Machine Mart were selling Allwins now. Must try and get to one of their 'VAT Free' nights.

[woody]

Hi did anyone read my post on the experimentation route?
The proof is out there....

Happy New Year to one and all

Cheers

Woody

[Bent Copper]

Yes I did. Unfortunately, I can't do the experiment myself because I don't think any skill is involved, so by definition, I wouldn't be able to play 'skilfully'.

Perhaps a better idea would be for Pennymachines to demonstrate his alleged skill to us all at the next auction on a Bryans Tencup, to show that he can come up with results consistently better than random. He's already said that the Variable Pressure control is ineffective because he can sense the pressure on his finger and automatically compensate for it, so that shouldn't pose any problem for him. I would certainly like to see these remarkable Allwin skills for myself.

[john t peterson]

Who says the internet is a "dumbing down" of intellect? I find this discussion facinating. My suggestion is to use the next auction as a test plaform for the hypothesis that allwins are skill more than luck (or phrased the other way around if you like.) Several standard allwins set on free play will be set up and the contestants will be allowed a specific number of shots, say 100. All winning shots per player will be totaled to see if some are better than others. Those in the "luck only" category will prevail if there is no statistical difference between the players. If one or more is able to beat the machine significantly more than the other players, then there is some level of skill involved. Just for the record, my money's on Pennymachines!

[pennymachines]

Congratulations to Bent Copper for starting the topic and to all who've chipped in. It's made some of us think a bit about our "simple" games and if it's achieved nothing else - it looks like being the first discussion to run over 4 pages!

Two points need clarifying judging by Guest's recent comments:
1. I have sworn never to abuse my super-human powers in the pursuit of personal gain. Besides, it would ruin my argument if I claimed special skills. My point is that any reasonably adept player can increase their winnings through skill.
2. Apart from on a few over-generous allwins, I don't claim that skill lets anyone win (i.e. profit in the long term) - merely that it allows them to improve the odds.

I would be very surprised if it makes any real difference where the ball lands, or even that it's possible to 'aim' the ball with any degree of accuracy. For instance balls that hit the right-hand cups are more likely to bounce left and vice versa. So balls will tend to bounce towards the other cups wherever they land.

Whether the ball can be "aimed", our first point of contention, can't be resolved by argument, but it seems logical that the nearer to the centre-top the ball starts, the greater its chance of landing in a cup on its downwards decent. I'm not sure why you say that a ball hitting the right-hand cups is more likey to bounce left. Surely it's nearer 50% left or right. In which case it has a great chance of losing as it falls out range of the cups.

However, as I said at the start, multicups are the hardest case. I feel they're skill games but suspect the degree of skill is so small it would require many plays before the statistics would show anything. On the other hand, it wouldn't be my first delusion...

Without revisiting the football analogy - my point about these other "skill" games was that they're subject to similar chaotic variables. Again it's a matter of degree - I don't say darts, football and allwins are equally skilful games, although in view of England's penalty kick record some of us may prefer to think so.

Now that the gauntlet has been thrown down you won't be surprised when I start quibbling about the rules. The first bone of contention will be over which allwins to use in our experiment. I don't think your suggestion of a small multicup is entirely fair, BC, in view of what I've said already. Did you choose this because you're beginning to doubt your position? I suspect you'd object equally if I asked for my Gapwin as the easiest on which to demonstrate skill. Why not even things out by testing both? But let's put this question aside for now. Demonstrating my "alleged skill" wouldn't support the claim that any reasonably adept player can increase their winnings through skill, so I prefer John's suggestion that lots of players should enter the test. Besides, it sounds more fun.

If one or more is able to beat the machine significantly more than the other players, then there is some level of skill involved.

The problem is, this will almost inevitably occur whether or not there was any skill involved (as it would if we played bandits). The Skills and the Lucks would just end up arguing over the statistics.

Woody's original idea sounds better - comparing the results of a number of skilful shots against an equal number of random shots. An improvement on this would be to compare aiming to win against aiming to lose. After all, if there is some skill involved, it should be at least as easy to deliberately lose as to win. The experiment should be confined to Skills (players who believe there is skill involved) otherwise those cynical Lucks might deliberately use their skill to skew the results in the opposite direction, just so they could win the argument!

Sorry I gave away that tip about the Machine Mart allwins. Looks like they've all been snapped up in the New Year sales.

[Bent Copper]

Congratulations to Bent Copper for starting the topic and to all who've chipped in. It's made some of us think a bit about our "simple" games and if it's achieved nothing else - it looks like being the first discussion to run over 4 pages!

To be honest, I think we're in danger of overdoing it now, but it's quite fascinating to have the opportunity to try and analyse these much-loved games in a learned and scholarly fashion!

Whether the ball can be "aimed", our first point of contention, can't be resolved by argument, but it seems logical that the nearer to the centre-top the ball starts, the greater its chance of landing in a cup on its downwards decent.

Ah yes. It seems logical. That could well be the Allwin playing tricks on you again, as it is been designed to do. Don't take anything for granted.....

I'm not sure why you say that a ball hitting the right-hand cups is more likey to bounce left. Surely it's nearer 50% left or right. In which case it has a great chance of losing as it falls out range of the cups.

I don't think it is an equal chance of bouncing left or right. If a ball hits the top surface of a cup from the left, it will bounce off to the right. But there are only 2 narrow top surfaces. If the ball hits the much larger vertical surface, it will bounce back to the left. This is a simplification to try and explain my point, but I think there's a much greater incidence of the ball bouncing back towards the middle of the playfield rather than skimming across the top of the cup and bouncing out.

I also think a similar situation may exist when a ball hits a gallery off-centre (ie balls will tend to bounce towards the centre), but I'm not claiming that to be the case. It's just a possibility that hasn't been considered. If that is the case, then it makes it less important where the ball first strikes the gallery, as it will always tend to bounce towards the side with the greater number of cups.

However, as I said at the start, multicups are the hardest case. I feel they're skill games but suspect the degree of skill is so small it would require many plays before the statistics would show anything.

Well that's the point I made earlier. There comes a point when any skill is so overwhelmed that it becomes insignificant, and therefore (to all intents and purposes) a game of chance.

Now that the gauntlet has been thrown down you won't be surprised when I start quibbling about the rules. The first bone of contention will be over which allwins to use in our experiment. I don't think your suggestion of a small multicup is entirely fair, BC, in view of what I've said already. Did you choose this because you're beginning to doubt your position?

No, not at all. Just previously you said that skill was involved in multi-cup Allwins (although you now seem to have back-tracked a bit) so I was just giving you the opportunity to demonstrate your point.

I suspect you'd object equally if I asked for my Gapwin as the easiest on which to demonstrate skill.

Yes, because this discussion is about conventional Allwins, not specials. I am shocked that you would want to cheat by using a special Allwin with a large open centre target. Did you choose this because you're beginning to doubt your position?

But let's put this question aside for now. Demonstrating my "alleged skill" wouldn't support the claim that any reasonably adept player can increase their winnings through skill, so I prefer John's suggestion that lots of players should enter the test. Besides, it sounds more fun.

Yes, and much safer for you too!

[badpenny]

Sorry I gave away that tip about the Machine Mart allwins. Looks like they've all been snapped up in the New Year sales.

I missed that, do you mean the shop that sells garden shed versions of lathes and compressors and sends me "manager's special offers" every fortnight? and all because I once bought a very useful box of split pins from them.

Incidently I used all the different versions of SEARCH on this site to try and discover the original quote and couldn't even find the one posted yesterday. I imagine that was my fault and urge you all to be grateful that when it comes to my hap hazard attempts at flicking swithes and pressing buttons I never succeeded to a career in the nuclear industry ......... also if you're looking for someone to indiscriminately fire ball bearings around a steel track just write "Search" above the trigger then come back and collect me after a day or two.
Badpenny

[bryans fan]

I would personally like to thank PennyMachines and Bent Copper for this lengthy debate, you have proved to my wife that there are people far more obsessed with slot machines than me!
What you two are doing up in the early hours is beyond me!
For what its worth, I have already voted skill.
I have found that with my allwins my sucess rate improves with practise. We have a 10 year old visitor who can win chocolate from a R&W allwin seemingly at will.
Incidentally he can virtually empty my circle skill of candy sticks until his mother puts a stop to it!

Thanks for a lively debate to all who have contributed. Nuff said?

PS I always get my machines from Screwfix

[pennymachines]

To be honest, I think we're in danger of overdoing it now

I suspect we overdid it on the first page, but what the hell!

previously you said that skill was involved in multi-cup Allwins (although you now seem to have back-tracked a bit)

On the contrary, I forward-tracked - in my first post I said a multicup was an example of an allwin where "luck clearly predominates" and later revised this to "I'm not certain a multicup is pure luck". That's the position I still hold. I'm not strongly convinced either way. If there is skill, I would expect it to take much longer to demonstrate.

In your first post, you made the bold claim that the allwin game "is just as random as a one-arm bandit." In my first post I said there were some allwins in which luck predominates and others in which skill is significant. So I could accuse you of back-tracking by wishing to disqualify any allwins I propose as skill games on the grounds that they're "special cases". Unlike you, I made no claim about allwins in general. All the games I allege are skill were common in arcades and the first task of a skilful player was to home in on them while ignoring the others.

Yes, and much safer for you too!

I didn't realize I was in danger. Perhaps I will have to call upon those super-powers!

Badpenny - put Machine Mart into the Forum Search, select "Search for all terms" and "Display results as: Posts" and you should get a full list of relevant posts. The full company name was Amusement Machine Mart and some of their allwins were Extrawin, Crackerjack, Time Limit, Win-A-Race, Beat the Clock, Double Your Win and Payola, any of which would make good test cases for the skill challenge.

Chaos Theory cropped up in a book I was reading last night, this time applied to billiards:

Suppose we forget about the effects of air resistance and friction that stop the ball moving on a real table. If we could hit the ball with absolute precision, we could, using Newton's laws of motion, predict the subsequent position and speed of all the balls exactly.... Suppose we could know the starting state even as well as quantum theory allows. This would then enable us to reduce our uncertainty about the cue ball's position to an accuracy billions of times smaller than the size of an atomic nucleus. Yet after only about fifteen collisions with other balls, this infinitesimal uncertainty expands to the dimension of the entire table.

[Bent Copper]

In your first post, you made the bold claim that the allwin game "is just as random as a one-arm bandit." In my first post I said there were some allwins in which luck predominates and others in which skill is significant. So I could accuse you of back-tracking by wishing to disqualify any allwins I propose as skill games on the grounds that they're "special cases". Unlike you, I made no claim about allwins in general. All the games I allege are skill were common in arcades and the first task of a skilful player was to home in on them while ignoring the others.

I didn't see the need to mention any particular Allwin in my original post, but perhaps I should have done. It was designed to provoke discussion (and succeeded). What I said was there was no skill involved in playing "an Allwin". I didn't say that there was no skill involved in playing "any Allwin that's ever been invented" which is your mistaken interpretation. It's obvious that the discussion has to be confined to generalities, otherwise anybody could just come along and say "I knew a rare type of Allwin once where skill was involved, so I say all Allwins are games of skill" (In fact this is what a number of people have said!) This is such a ridiculous interpolation that I didn't expect anybody to propose it as a serious argument, and I didn't think there was a need for me to specifically rule out 'special' Allwins in my original post.

So I am sorry if my original post wasn't worded specifically enough. By "an Allwin" I was referring to the common configurations similar to a Playball or a 10-Cup (the type of Allwin that you would draw if somebody asked you to draw an Allwin). Almost every Allwin manufacturer made these types of machines, and this is what people would consider to be the 'standard' Allwin configuration. I suppose the Bryans versions would be the Ten-cup and the Five-win.

I also made the mistake of not defining what I meant by skill and luck. It has become apparent that people's definition of a skilful game varies; something which I hadn't anticipated. I agree that there may be a small amount of perceived skill involved, but I think this is completely overruled by so much randomness in the machine that any skill element becomes negligible and can be discounted. Other people have said that providing there is some minute skill or influence involved, however insignificant, then it must be a game of skill. I don't agree with that as a definition of skill. You could use the same argument to say that tossing a coin was a game of skill. Strictly speaking, Yes it is, but not in any real or practical way. I think it is the same with an Allwin. Clearly the definition of a game of skill is a subjective one. I suppose what I should have asked is: "Does luck or skill predominate?"

Chaos Theory cropped up in a book I was reading last night, this time applied to billiards:

Suppose we forget about the effects of air resistance and friction that stop the ball moving on a real table. If we could hit the ball with absolute precision, we could, using Newton's laws of motion, predict the subsequent position and speed of all the balls exactly.... Suppose we could know the starting state even as well as quantum theory allows. This would then enable us to reduce our uncertainty about the cue ball's position to an accuracy billions of times smaller than the size of an atomic nucleus. Yet after only about fifteen collisions with other balls, this infinitesimal uncertainty expands to the dimension of the entire table.

And what did they say about Allwins?

[Bent Copper]

And I have to admit that I couldnt drop the ball in the same position with any certainty at all. The reason is that there seems to be a very fine line between the ball dropping to the left and dropping to the right and I don't think it's possible to operate the machine with that much precision just by feel alone.

I think you've hit on an important point there. It was always my impression that the difference in momentum between the ball going to the left or going to the right of the playfield was much smaller than you'd expect. But this is something I can't prove or quantify in any way.

I first noticed this when I once played a U-Win. It became apparent to me that most balls missing the left-hand lose hole would have sufficient momentum to carry them right over the 'U' and into the right-hand lose hole, even though this didn't seem logical. Bryans obviously knew about this effect and very cleverly designed the shape of the U to maximise it, but I think it is present on all Allwins.

It's probably this same effect that is responsible for many more balls dropping on the right-hand side of a gallery than on the left-hand side. It seems that most of the balls that make it up the left-hand side of the spiral have sufficient momentum to either do another circuit, or fall just short and land on the right-hand side. Only rarely does a ball just make it over the track and land on the left-hand side (it happens of course, but far less often). Why is this?

Perhaps it is because the ball is travelling from left to right when it's going over the top, but the momentum could just as easily run out when the ball is on the left. Perhaps there's a different amount of friction at the top and bottom of the track and the ball loses most of its momentum at the bottom, and hardly any at the top. Perhaps it's due to the effect of gravity, which has the opposite effect on the left and right side of the spiral. Whatever the cause, this right-hand bias means that the momentum has to be controlled to within very fine limits to 'aim' the ball anywhere other than on the right-hand side of the gallery or playfield.

[john t peterson]

"It is more effective to beg forgiveness than ask permission." Socrates.

[bryans fan]

My wifes` first question is always "where is it going?" So perhaps it would pay you to have an answer ready!! Now that`s skill!

[jimmycowman]

just ask the wife one question why do you want so many pairs of shoes?!?!? youve only got one set of feet

[Bent Copper]

Mr Kirkup said: "If there is an element of chance (e.g. the shuffling of the cards), I am satisfied that it is gaming." - the same reason that allwins are currently adjudged to be games of chance.

The Gaming Board's opinion is as extreme as some of the contributors here, but in the opposite direction.

While the Gaming Board's attitude is that if a game is only 1% luck and 99% skill, it is a game of luck; many people here seem to think that if a game is only 1% skill and 99% luck, it is a game of skill.

Personally, I think both of these extreme views are as nonsensical as each other - although it's nice to have the law on my side (for a change).

Of course, any legal definition has to be objective and unambiguous, so there is a need to adopt the extreme view, but unfortunately, this doesn't tell us anything about the mechanics of Allwins!

[margamatix]

I think Mr Bryan built his business and made his fortune from people who thought an Allwin was a game of skill.......I have also known people (and I'm sure this is going to ring a bell) who used to claim that pulling down the handle on a mechanical one-armed-bandit VERY VERY SLOWLY and then giving it a sharp tug just before tipping point would influence the chances of winning.

All successful amusement machines work on the principle of "Oooh, I was so close, I'll just have another go because I'm bound to win next time"