Back in July 2014, whilst on a day trip to Whitby, I ate at the Angel Wetherspoons’ pub. I sent this tweet.
@jdwtweet I generally love your food but this at Angel Whitby is a little stingy!! pic.twitter.com/YRtkWct19O
— Doug Paulley (@kingqueen3065) July 16, 2014
Wetherspoons responded by pulling the CCTV of our meal and interviewing the waitress. They indicated that the portion size was within one standard deviation of the mean of their standard so they were content with the size of the portion. They accounted for my disapproval with the observation that I had been eating for precisely 2 minutes and 17 seconds when I took the photo, and stated that the waitress had testified that I had not indicated any displeasure to her at the time of the meal.
I was impressed with this commitment to customer satisfaction, so when I attended The Corryvreckan (Wetherspoons’ pub in Oban) whilst on holiday last week, I decided to support their analysis with the provision of data from another sample. This is therefore a comparative study of the size of Wetherspoons’ Steak and Kidney pudding meal.
The diameter of the meal is approximately 18cm, on a patterned plate of approximately 25cm. The surface area of a plate of diameter d is approximately (πd^{2})/4, or in this case 490cm^{2}. Of that, approximately 250cm^{2} was obscured by food or by the gravy pot, that being an occlusion of approximately 51% of the plate. This appears to be roughly equivalent to that of Whitby in 2014; though I note that the practice of providing a gravy boat may give the impression of more food than previously.  
The chips appear to be distributed on the plate in a pseudorandom distribution. The average depth of food on the plate is therefore difficult to estimate, but is perhaps a mean of 1 or 2 centimetres from the deepest point.  
The pudding varies in diameter between 7cm at its “base” (the top in this picture) to 10cm at its “top” (the bottom in this picture.) The pudding is approximately 6cm in height. Using the reasonable approximation of a cylinder of diameter 8.5cm, its volume can therefore be approximated using the formula volume=(πd^{2}h)/4=340cm^{3}. The density of cooked ground beef is approximately 1.03gcm^{3}, essentially indistinguishable from the density of distilled water at standard temperature and pressure (1gcm^{3}), so I estimate the mass of the pudding is approximately 340g. Comparative research of other single portion steak and kidney pies reveals that this is within an order of magnitude of expectation.  
There were precisely 30 chips, varying in length between 2cm and 12cm, with a median length of perhaps 8cm. There was therefore approximately 2.4m of chip on the plate.  
The average thickness of each chip was 0.64mm. Given the presence of some outliers with tapered ends, I am estimating the total volume of the chips on the plate as being 240cm x 0.6cm x 0.6cm or 86.4cm^{3}. Fried potatoes have a density of 449Kg/m3, or approximately 0.5gcm3, so I estimate the total mass of chips to be approximately 43g.
Research indicates that the average portion of cooked chips is 200g, and that a few chips either way can make large cost differences. I frankly suspect some scrimping here. 

There were 169 peas. They averaged 0.45cm in diameter. The volume of a sphere of diameter d is (πd^{3})/6, so each pea measured approximately 0.047cm^{3}. The total peaage was therefore approximately 8cm^{3}.
The NHS states that the “five a day” portions of veg can include “three heaped tablespoons of cooked vegetables”. A heaped tablespoon is 30cm^{3}. This is therefore about a third of a portion of peas on that plate. The density of cooked peas is 0.68gcm^{3}. The mass of peas was therefore approximately 5.5g. 

The gravy boat is a new addition since 2014. The depth of the gravy is approximately 3.5cm.
The average individual portion of gravy is approximately 50cm^{3}. For the gravy boat to hold that amount, it would have to have a surface area of perhaps 14cm^{2}. I estimate that the surface area of this gravy portion exceeds this and therefore we are on the up. But not all the gravy got eaten, as I was not furnished with a spoon. 

The total mass of the pudding, the peas and the chips was therefore approximately 390g.
The average eating rate varies substantially by individual, food type and circumstances but is approximately 100g per minute. This meal would therefore take the average person approximately 4 minutes to eat. Of course, because I was being sarcastic and pissing about with a camera and a ruler, it took me substantially longer. 

My blackcurrant and soda was approximately 11cm in height and the glass was approximately 5.5cm in diameter. Its volume = (πd^{2}h)/4 = 260ml, or just less than half an imperial pint.  
But, of course, some of that was taken up by ice cubes. There were 5, with an average size of 1.5cm. The total volume of ice was therefore 5 x 1.5 x 1.5 x 1.5 = approximately 17ml.
In approximately half of cases, restaurant ice has over 1,000 colliform bacteria per cube (i.e. faecal bacteria) and is thus more contaminated than toilet water. On average, there was therefore probably 2,500 poobased bacteria in the ice cubes in that glass. 

The receipt varied from 19cm in length to 20.5cm, at a width of 8cm. The total surface area of the receipt was therefore 164cm^{2}. At an assumed weight of 58gm^{2}, the paper weighed approximately 9mg – or approximately a billionth of the mass of this European oak tree. 
Conclusion
I hope that this comparative study of the dimensional composition of my meal is to the exacting standards of Wetherspoons and contributes to their body of statistical analysis of their meal – and I look forward to their prompt analysis as to whether their Oban staff complied with Wetherspoons’ evil corporate pennypinching controlfreak bollocks.
2 Responses to “The Consistency of Wetherspoons Portion Sizes”
Sorry, the comment form is closed at this time.
I think you need to investigate packing of spheres, in particular, the packing of spheres in a pseudospherical open container (e.g. a “Spoon”). Mathematically, this is a complex matter:
https://en.wikipedia.org/wiki/Sphere_packing
I also think that an investigation into the range of sizes of peas is in order: variation within the pea world may not be obvious to the human eye but is an important aspect of pea volume estimation.
Natural variation in pea size is something that botanists have studied, and there is an extensive literature on this e.g.:
http://www.ncbi.nlm.nih.gov/pubmed/17727416
However for the purposes of Wetherspoon evaluation, might I respectfully suggest an empirical study? It would require some peas, and a tablespoon. My guess is that three heaped tablespoons of peas is not as large as you estimate, and that the portion provided does fall with the government recommended 5aday range. I look forward to reading your followup post on the empirical distribution of peas when measured using a tablespoon, in terms of mass, volume, and number, both within and between spoonsamples.
Sometimes I really worry about you Doug!