Halloween Masks

Friday 30th October 2015

A quick one this week, if you're going to a Hallowe'en party this week I'd totally recommend these awesome masks from Wintercroft, a Cornish design company. Here are some examples from their website:

They're £4.50 each and are emailed to you as a PDF. You cut them up, glue the templates onto some card and assemble them. There's a video on Youtube showing how to make one.

This one took me about two hours to make, and it's great! Okay, so mine's not as good as some of the examples above :)


Recent posts

How do casinos make money?

Friday 23rd October 2015

I've been obsessed with poker for about five months now. I've just this week completed my goal of turning an initial £10 stake into £100! There were two things that originally got me interested in it.

Firstly, unlike many other gambling games, it involves a degree of skill. I thought that if I learned the maths and probability behind it, maybe I could do well.

Secondly, playing it in a casino doesn't involve playing against the house. And as we all know, the house always wins. For example Blackjack and Roulette incorporate rules specifically designed to give the casino an edge of the players, and slot machines have a fixed percentage that they return over time.

I didn't really understand how it all worked, so I've done some reading and here's what I've learned. It turns out I was wrong!



The house edge in Blackjack comes mostly from the rule that if the player busts then they lose - it doesn't matter if the dealers subsequently busts too. This happens often enough to make a bit of money for the casino every time it does.

This means that even a good Blackjack player will lose on average 1% of the their stake over time, but 1% is pretty good compared to some other casino games.



In the case of Roulette the house edge comes from the green 0 pocket on the wheel.

Using European Roulette as an example, there are 37 pockets that the ball can fall into. If you put £1 bets on all 37 numbers you'd be guaranteed to win £35 every time the ball rolled, losing you £2 per go. The house edge here is 2.7%.

Or let's say you place £1 on black and £1 on red, so you'll win £2 back most of the time. But every 1 in 37 times (on average) the ball will fall into the 0, so over time you're losing 2.7p each time the ball rolls. The house edge is still 2.7%.

American Roulette has a 0 and a 00 pocket, which increases their edge to 5.3%.

To be honest I don't know why anyone ever plays Roulette, but many people over the years have told me about a mathematically proven system that beats the house.


The Martingale System

The idea is simple - you place a bet on red (or black if you prefer) and each time you lose, you double your stake. This is formally named the Martingale System

As an example:

  • I bet £1 on red
  • If red comes up, I've won £1
  • If red doesn't come up, I bet £2 on red next time
  • If red comes up, I've won £1 (£4 in winnings minus £3 in stake)
  • If red doesn't come up, I bet £4 on red next time
  • If red comes up, I've won £1 (£8 in winnings minus £7 in stake)
  • If red doesn't come up, I bet £8 on red next time
  • If red comes up, I've won £1 (£16 in winnings minus £15 in stake)
  • and so on...

Eventually I'm bound to win, because red has to come up at least once!

The first reason this doesn't work is that no matter how much you bet, you only win back your original stake. So you'll need to make your original stake really large.

Say we started off with a £1,000 bet on red, eventually we're going to win back £1,000 when red comes up, even if it takes ten spins. But if it takes ten spins, then by spin ten we're betting £512,000 just to win £1,000! If you've got half a million quid lying around, are you really going to risk all of it to win a grand? And what happens if it takes twenty spins? Pretty unlikely, but you'll need just over £1bn in case it happens.

The second reason this doesn't work is because of the 0. One time in 37 (on average) the little ball is going to fall into the zero pocket and you've got to start all over again. It's pretty unlikely to happen once (2.7% likely in fact), but in our ten-spin example it's not looking good for you.


Slot machines

This, I discovered, is where the really big money comes from. Slot machines are programmed to return a fixed percentage back to gamblers over time.

This is typically around 80-90%, but that's over millions of spins, and you're not guaranteed to get anything like that back.

In this example photo that I snapped in a pub the other day, it's 74%. Those are pretty terrible odds.

Taking Nevada in 2009 as an example, poker brings in around 1.6% of the total gaming revenue for the state.

Compare this with 4.7% for Roulette, 13% for Blackjack, and a whopping 51% for slot machines!

So let's avoid those too.



So I started playing poker online, thinking that I was being clever by not having the compete against the house. But pretty quickly I started to wonder how websites like Pokerstars was making money out of all this. And that's when I discovered rake.

Rake is the commission taken by a poker room (real or online) from a game. It ranges from 2.5% to 5% depending on where you play, and is usually expressed in a tournament buy-in as £X + Y where X goes into the prize fund and Y is the rake that the poker room take for organising the tournament.

For example, I entered a 6-handed tournament on Pokerstars last week that was advertised as a $1.50 buy-in (I know, big spender!) which the details listed as $1.40 + $0.10 entry fee, with a prize pool of $8.40. So Pokerstars are making $0.60 each time one of these tournaments start.

To give you an idea of how much money they're making out of this, they recently celebrated their 100 billionth hand, and most of their tournaments cost more than $0.50 to enter. Last year Pokerstars was sold for $4.9 billion in cash.

It turns out that rake is the major factor stopping the average player from winning money at poker. One poker forum post said "rake isn't the most important thing, it's the only thing".

Although it looks like they're giving you free drinks and food, in a live casino you're actually paying something like $15 an hour to play.

Of course you are, and I can't believe I ever thought differently.

Learning this has really changed my opinion of poker - I'd much rather play a home game with some friends, because I know that no-one's taking my money apart from my mates!



Recent posts

Me and my genome

Friday 16th October 2015

I just got my DNA sequenced. We're living in the future.


A very short history of DNA sequencing

1995 - A few months before I started university the first free-living organism's DNA (Haemophilus influenzae) was sequenced at a cost of hundreds of millions of dollars. This bacteria has 1.8m base pairs.

2001 - A couple of years after I left university, the Human Genome Project published the first complete human genome at a cost of $3 billion. Human DNA has around 3bn base pairs.

2015 - Fourteen years later I can get my DNA analysed for £125.


Sorry, what?

A company in the States called 23andme offers personal genome testing. You spit in a test tube and send it to them. They look at your DNA to identify your ancestry, assess your risk of genetic diseases, predicted response to drugs, and check for inherited conditions. They started in 2008 but only began offering the service in the UK at the end of last year.

Ever since I found out they were offering the service in the UK I've wanted to get it done, and writing it up here seemed like a good excuse to do it! This blog is the best idea I've ever had.

So what have I learned?


Photo by Steven Coles, https://www.flickr.com/photos/stewf/270942458

I am the 0.1%

No real surprises here, I'm 99.9% European.

The remaining 0.1% is from Yakut, a part of northern Russia the size of India with less than a million inhabitants. This explains why I've always loved playing the board game Risk!

An estimated 2.8% of my DNA is from Neanderthals, compared to a European average of 2.7%. Neanderthals are the closest evolutionary relatives of modern humans, who went extinct about 40,000 years ago.


Genetic Risk Factors

This is the bit I'm really interested in, I'm not that bothered about how much DNA I share with Neandethals. 23andme looks at sections of your DNA that are associated with an increased risk of a variety of diseases. If you come up as a higher risk, it doesn't mean that you're definitely going to get the disease, but you might want to look into it further.

23andme gave me typical risk for nine genetic risk factors including Parkinsons, but it did flag up higher risk for two diseases that I wasn't aware of. It turns out that the rest of my family are well aware of one of them, so I'm looking at ways to avoid it.

For the other genetic risk I've registered for clinical trials with the National Institute for Health Research, to see if I can help find a cure in case I get it! I'm not really concerned about this, it's a late-onset disease and there's no history of it in my family who have all survived way past the age it normally appears.


Drug response

This section of the report gives me the all-clear for expected responses to drugs like Warfarin and Abacavir which can cause problems if you're sensitive to them.

It gave one abnormal response, I'm likely to metabolise a certain kind of medicine more quickly than normal. So if I ever do need to take it, I can let my doctor know that I might need a higher dose than most people.

Given how much doctors seem to hate self-diagnosis and information gleaned from the internet, I can only imagine my doctor's response to this.



Inherited conditions

A clean sweep here for 45 inherited conditions. To be honest I'd probably already know if I'd inherited most of them, like cystic fibrosis or the delicious-sounding Maple Syrup Urine Disease.


Guess Who?

The last section is fun - do my genes match me? According to 23andme:

  • I'm likely to have straight-ish non-balding blonde hair and blue eyes
  • I drink an average amount of coffee and metabolise it more slowly than average
  • I can taste bitter things, I have typical odds of disliking coriander and can smell asparagus in my urine after eating it
  • I have a lower than average preference for sweet foods. 



Absolutely fascinating, I'm really glad I did it. If you're interested here's a link - let me know what you discover!



Recent posts


Friday 9th October 2015

I've always assumed that Astrology is complete rubbish, but I've never actually paid any attention to it. So I thought I'd do some experiments to see if it stands up to any kind of close examination. Here are the results of my experiment.


1. Does my star sign match my personality?

I am apparently a Sagittarius. I assumed this meant that I have four legs and I'm good at archery, but here are a list of traits of Sagittarians according to six different websites.

I asked Kat to score how many she thinks apply to me, I'll leave it up to you to decide which she chose.

Website #1 Website #2 Website #3 Website #4 Website #5 Website #6
  • kind-hearted
  • open-minded
  • enthusiastic
  • quick wit
  • thirst for knowledge
  • mental acuity
  • strong
  • stubborn
  • loves finer things in life
  • appreciates beauty
  • feelings buried deep
  • generous
  • enthusiastic
  • passionate
  • adventurous
  • direct
  • performs well under pressure
  • unemotional
  • generous
  • dependable
  • down to earth
  • patient
  • independent
  • persistent
  • high-energy
  • eager
  • idealist
  • passionate about justice
  • equality-promoting
  • practically grounded
  • creative
  • reasonable
  • perceptive
  • sensitive
  • compassionate
  • self-doubting
Score: 6 Score: 3 Score: 3 Score: 3 Score: 2 Score: 4


In fact, only three of these are traits for Sagittarius, the ones in the columns with the dark backgrounds. #2 and #4 are Taurus and #6 is Pisces. So on average 61% of the Sagittarius traits match, against 55% of the traits for the other signs. Not a great start, but it's a small sample size so I'm not sure we can read too much into it.

You'll note though that are is some degree of consistency between the traits for Sagittarius: enthusiastic and passionate both appear on two different lists, and generous appears in both of the Taurus lists.

Result: inconclusive


2. Do these personality traits agree with other people?

I know three people who share my birthday, so they should share these Sagittarius traits too. I emailed them a list of these traits and asked their partners to tick the ones that applied. I didn't tell them what it was for. Only one replied (fair enough, it was an odd request): Mr H matched 72% of the Sagittarius traits and 33% of the non-Sagittarius traits, which is a much stronger match than me.

Result: promising



2. Do my horoscopes agree with each other?

This seems like a fairly basic expectation of a horoscope. If they're predicting my future, they should at least agree on what that future is. I Googled horoscopes, opened the top six websites and found today's text for Sagittarius. I've summarised them into bullet points:

  • horoscope.com - You'll get an angry call from someone far away
  • Yahoo! - You're able to work more quickly and efficiently.
  • cainer.com - What's positive, this weekend, may well be possible, if you believe!
  • horoscopes.co.uk - Together with someone close, you could be planning for the future.
  • Elle - You could blow a situation way out of proportion
  • Daily Mirror - Today you will recognise when someone is feeding you false leads.

There's 0% consistency between these, there's not one common theme among them without stretching the words to breaking point.

I've also just noticed that Russell Grant provided the predictions for both Yahoo and the Mirror website, which don't even match up to each other.

Result: fail



3. Does my daily horoscope predict my day?

If they really are true, then I should be able to see correlations between what my horoscope predicts and what actually happens in my day.

In the examples above for today, I would argue that the last one about being fed false leads does make sense! But that means if it's false it's true then it's false and we're in some kind of mind-melting paradox.

I'm giving Yahoo Russell a chance to predict my future. Here are my predictions for a week, and my experiences:


  • Russell predicts: You're able to work more quickly and efficiently.
  • My day: No, I spent most of the day elbow-deep in a database and hardly achieved anything.


  • Russell predicts: You'll be asked to make a personal sacrifice for a loved one. 
  • My day: I agreed to go to a vegetarian restaurant for lunch, does that count?


  • Russell predicts: People are receptive to your particular brand of genius. 
  • My day: Thanks for the compliment, but no.


  • Russell predicts: You're determined to reach your goal, even if it means going above and beyond the call of duty.
  • My day: This doesn't sound like my Monday, can you be a bit more specific please?


  • Russel predicts: Your income is steadily improving, although you might not see a discernible change yet.
  • My day: The first part is true, but I have noticed it.


  • Russel predicts: Attending a religious, educational or cultural gathering will put you on the path to romance. 
  • My day: Ooh! But no.


  • Russel predicts: A financial shortage is putting a crimp in your social life.
  • My day: Utterly wrong, today I calculated that we're debt-free for the first time in 20 years!

Result: fail



And there you have it. Unsurprisingly this isn't the first ever scientific study into astrology, this study from 2003 found that astrological effects for 110 variables were not detectable in 2,101 time twins born 5 minutes apart. [...] The effect
size due to astrology was measured as 0.00 ± 0.03. I'm not sure what the units are there, but I don't think it matters much.

But I doubt it's going to change anyone's mind either.

What star sign are you, and has a prediction ever come true? Leave a comment below.



Recent posts

Open cola

Friday 2nd October 2015

The recipe for Open Cola, an open-source recipe for a Coke-type drink, came out in 2001. I've always wondered what was in Coke so thought I'd have a go at making some with Ash and Pete.

Here's what happened.

First of all we got all ingredients together. This was no mean feat in itself, and cost about £40 - but this is enough to make several batches. We sort of followed the recipes on Wikipedia and Cube-Cola. Most of the essential oils came from an online supplier, with everything else coming from eBay (I love eBay).

Flavouring formula

  • 3.50 mL orange oil
  • 2.75 mL lime oil
  • 1.25 mL cassia oil
  • 1.00 mL lemon oil
  • 1.00 mL nutmeg oil
  • 0.25 mL coriander oil
  • 0.25 mL neroli oil
  • 0.25 mL lavender oil
  • 10.0 g food-grade gum arabic powder
  • 40 mL water (the Wikipedia recipe says 3ml but this really didn't work for us)

Put on some gloves and goggles, the essential oils are pretty fierce. Measure out the oils and mix together, as shown below left. Mix the gum arabic powder with the water until it forms a cloudy liquid, and then add the oils.

You'll need to blend these together into an emulsion, which can take several minutes in a food processor or with a whizzy stick. Eventually it will form a cloudy emulsion as shown in the bottom right. You can tell it's an emulsion because if you put a drop into a glass of water, it won't separate into a oil layer on the top of the water. Don't leave it in the food processor too long because the oils start to attack the plastic.

We (okay, I) made a right mess of this step because we didn't dilute the gum arabic with enough water, so we wasted quite a lot of the mixture. We ended up with 30g of the emulsion when it should have been nearer 60g.

The mixture smelled absolutely incredible but it didn't taste too great - very very bitter.


The two recipes we were following deviated quite widely here, so we made a new middle path.

  • 1.5kg plain granulated white table sugar
  • 750mL water
  • 30.0 mL caramel colour
  • 17.5 mL 75% phosphoric acid (particularly nasty, watch out)
  • 30.0 mL flavouring formula
  • 1g caffeine tablets, crushed up, diluted with 10ml of water and passed through a coffee filter

Put the sugar and water in a pan and heat, making sure it doesn't boil. Add the caramel, phosphoric acid (carefully!) and caffeine solution to the flavouring formula and stir carefully.

My pH sticks said that the mixture was about pH 1, so don't drink it.

Once the sugar has all dissolved in the water, adding the flavouring and stir, then allow it to cool. We made just over two litres of the concentrate, enough to make 14 litres of cola. This is what the concentrate looked like.


We diluted the mixture 7:1 with sparkling mineral water and tried it - absolutely delicious! We had a bottle of Coke to compare it with, and to be honest after our magical open source cola, the "real thing" tasted dull and lifeless.

I've since tried it with soda water (also nice) and with straight tap water and it's still great! I've had really good feedback from everyone who's tried it, although almost nobody has guessed what flavour it's supposed to be - especially since it's not brown.

So it's a winner in terms of taste, how about cost? If we hadn't messed up the first stage we would have had more like 28 litres of cola and enough ingredients left over to make another 2-3 batches. That's about 70 litres of cola for £40, which is about half the cost of buying it.


Recent posts

What's this about?

Hi I'm Mat and I'm addicted to new hobbies. I used to think this was a bad thing but now I'm embracing it.

Writing them all up in this blog encourages me to finish projects, and helps me keep track of which ones I've tried.

Also in Making things...

Halloween Masks

Halloween Masks

A quick one this week, if you're going to a Hallowe'en party this week I'd totally recommend these awesome masks from Wintercroft, a Cornish design company...

Also in Food and drink...

Apple peeler

Apple peeler

Need a bit of light relief after recent events? I've admired these apple peelers ever since I first saw one - if you've never seen them before watch the video below! They were first patented by D...