Email in profile - would love to connect.

Calculations for slot machine mechanics and payouts have been in Excel for a long time. There can be a LOT of complexity in these workbooks. Sometimes it’s tricky to debug - but what’s the alternative? Code is often hard to debug too.

Simulating results (Monte Carlo) is nice but having two sets of data for validation/checking against each other is nice.

I am not aware of any alternatives.

Hard problems or domains are unknowns. Working towards solving hard problems involves thinking through unknowns, which may or may not lead to understanding. An aversion to hard problems is an aversion to the unknown.

In the usual Excel Solver case, for a simplified example, I want to create a biased coin where getting heads doubles your money and tails loses your money, with expected payout 90% in the long run. The parameters to change are heads/tails coin weighting, to target a value 90%. A fair coin gives a long run payout of 100%. It turns out that if a biased coin hits heads 45% of the time and tails 55% of the time, we end up with this 90% payout. Excel solver can come up with these two values.

In this example, we can come up with these 45% and 55% values theoretically. With more complex systems, we may want to see the effects of changing a subset of parameters that have an indirect effect on payout.

For example, if we extend the “biased coin game” to instead be “win your money back, 2, or 3 times your wager each with some probability” on a heads result, there are more parameters (and many solutions!) to get to 90% payout. Changing the heads probability has a further, second-order effect on the payout. Using Excel solver it’s straightforward to fix some parameters (heads/tails probability) and allow others to change (1x/2x/3x odds) to get desired results (90% payout).

If we further extend this methodology for a biased coin game, we can end up with a coin game that pays with distribution exactly like a slot machine in a casino, though it would not look like one!

It’s great, but you have to be able to put your problem into Excel to use it which does not work for all use cases.

In cases where I’m trying to optimise something that doesn’t quite fit into Excel cleanly I’ll usually do some sort of hand-rolled Monte Carlo optimisation. This generally works for me in the types of problems I most often solve.

Comments URL: https://news.ycombinator.com/item?id=29365428

Points: 4

# Comments: 1

Good work life balance, no crunch time, and decent pay especially when compared to the video game industry.

I’d probably be able to transition to the video game industry, but I don’t really have a desire to.

As someone using Excel to do slot game mathematics calculations, sometimes the excel formulas used are very long, convoluted, and hard to grok when coming back to them. Many cells often exist as calculation cells only, used for intermediate steps which leads to even more logic complexity.

I’m excited to experiment with these to try and simplify some of the long, previously-deemed-necessary calculation methods.

https://bmm.com/

1. On StackOverflow learning more about how C# works under the hood to understand memory/CPU usage (newer to C#, used C++ in a past life)

2. Building/debugging a spreadsheet in Excel (not a software engineer by job description)

3. Debugging software issues on hardware by messing with the hardware

4. Thinking about how to code to solve the problem; drawing/writing things out, etc.; rather than just diving in head-first (need to do this more...)

Excel is great in my industry (slot machine game/math design). It is fantastic for doing game calculations and there are reasonable ways to do error handling/checking for correctness.

I could say the same about programming a web app without writing tests/spending some time on system architecture. It looks fine until it doesn’t work.

In both instances of development (I argue constructing an Excel workbook in my line of work is very similar to programming) there are ways to mitigate risks by doing things similar to “writing tests”.

People that have this skillset (knowledge of Monte Carlo methods for statistical analysis/verification) are useful in this sphere.

An example problem we deal with is as follows:

We have some arbitrary histogram (list of slot game awards and probabilities), and we wish to map this to a particular game of bingo (N ball calls out of M balls in total, 5x5 bingo card). The constraint here is that when you come up with a list of bingo patterns for awards, they are ordered in the sense that when evaluating the set of bingo patterns, from top-down, the first bingo pattern that matches is awarded and the evaluation stops.

Getting the optimal fit theoretically in the general case requires exponential time, as the chain of P(this pattern given (not-that pattern and not-other pattern and not-...)) grows with an exponential number of operations in the theoretical expansion. Each time a new pattern is fit to an element of the histogram, the probabilities of all remaining possible patterns to fit changes.

There are a few ways of tackling this problem in a more optimal way: If you can solve this with good enough accuracy in linear time with a good math/cs background, you’d be useful to any company playing in this space.