# Brewing by the Numbers | The Law of Averages Part 1

### Help Support Homebrew Talk:

The technical side of brewing has always been the most interesting to me. I enjoy the process (and the drinking), but the science behind it has always taken the limelight. I was reading the recipe for Biermuncher's Aberdeen Brown Ale (https://www.homebrewtalk.com/f67/aberdeen-brown-ale-newcastle-clone-ag-36912/) and he mentioned that Newcastle blends two different batches. More importantly, he mentioned how he opted to blend two recipes instead of making two batches. That really got my mind churning. Could that be applied elsewhere in brewing?
So, I decided to try a little experiment called "The Law of Averages." I took three different pale ale recipes from three respected and well-known brewers on HBT: Biermuncher, Yooper and EdWort. Here are the links:
Biermuncher's Six Shooter Pale Ale:
Yooper's House Pale Ale:
https://www.homebrewtalk.com/f66/da-yoopers-house-pale-ale-100304/
EdWort's Haus Pale Ale:
https://www.homebrewtalk.com/f66/bee-cave-brewery-haus-pale-ale-31793/
The experiment consisted of merging the three recipes and seeing what the end product would be. This is entirely experimental and I do not mean to infer that the final product will be better than the base recipes or that any of them are better than the others. This is pure curiosity...to the graphs!
The first graph shows the percentages of the three recipes: base vs. caramel. EdWort's Haus Pale Ale had the highest base:caramel ratio at 95:5. Biermuncher's Six Shooter had the lowest at 88:12. If you follow the numbers over to the last column, you'll see the "Ideal Grain Bill" column. The percentages in that column are averages of the other three recipes, which are 91.33% base malt and 8.68% caramel malt. Simple enough?
The second graph breaks it down even further. It breaks the base malt down into specifics. Yooper's House Pale Ale had the most even distribution across base malts with 50, 30 and 20% of the base malt being two-row, Vienna and Munich, respectively. (I also noticed the top row in that graph didn't format the numbers as percentages). EdWort's Pale Ale had the simplest breakdown with 80 and 20% of the base being two-row and Vienna, respectively. So, the averages of base malts leads to an ideal mix of 69.3, 20.3 and 10.3% of the base malt being two-row, Vienna and Munich, respectively.
Of course, it wouldn't be in-depth enough if I didn't delve into the caramel malt with the same veracity. The third chart breaks the crystal down into two categories: light and medium. This category was the most diverse with Biermuncher's Pale Ale being an 80/20 mix, Yooper's half and half, and EdWort's being completely light crystal. I know everyone has their interpretations of crystal/caramel malt, but I use 0-40L as light, 40-80L as medium, and 80-120L as dark. The averages led to a 77/23 (light/medium) mix. I am choosing to go middle of the road on both: Crystal 20 for the light and Crystal 60 for the medium.
Last, I analyzed the hop schedules. This last bit fascinated me because of the difference in schedules, which I thought would make for an interesting mix. Biermuncher used all late hop additions, Yooper used mostly late additions, and EdWort used an even spread. If a hop addition fell on the border of categories (0-15,15-30,etc.), I split the addition in half to account for both categories. For example, if it called for a 1 ounce addition at 15 minutes, I added 1/2 ounce to the 0-15 category and the other half to the 15-30 category. The averages led to roughly 50% of the hops for bittering and flavor, and the last half for aroma.
There were other calculations that I didn't make charts for because it wouldn't make any sense. Here were the others and how they contribute to the recipe I will be using.
Pounds of Grain Per Gallon: Biermuncher used 20.5lbs of grain for a 11.5 gallon batch. 20.5 divided by 11.5 gives roughly 1.78lbs of grain per gallon. I averaged all of the recipe lbs/gallons ratios for a final answer of 1.93lbs of grains per gallon.
Efficiency: Biermuncher's recipe was calculated at 72%, EdWort's at 75%, and Yooper's wasn't mentioned so I just went with 70%. Averaged out, it gives an average efficiency of 72%.
For the experiment, I'm going to make a one gallon batch with the following numbers derived from the averages of all three recipes:
OG: 1.051
FG: 1.014
IBU: 40.6
SRM: 7
Yeast: US-05
Grain Bill:
1.76lb Base
- 1.22lb 2-Row
- 0.36lb Vienna
- 0.18lb Munich
.17lb Caramel
- 0.13lb Crystal 20
- 0.04lb Crystal 60
Hop Schedule:
- .143oz at 60
- .044oz at 30
- .105oz at 15
- .264oz at 5
Part 2 of the article will be a review of the final product. Thanks for reading and feel free to apply this as you see fit. Cheers!

Very in depth. Great job with this. Will you end up doing this for other popular recipes in other categories also? For example IPAs, Belgian Wits, pilsners, etc.

Thanks! I am already doing the number crunching for other styles.

Wont this produce an average brew?

it might be too late now, but I'd think about using alpha-acid-ounce units for the hop addition, instead of just ounces. At least for the early hop additions. An oz of cascade at 60 min vs an oz of columbus will produce pretty different results. Perhaps if you do an IPA comparison, you could try it

I like this analysis and will probably use a similar method from now on when coming up with recipes (I kind of already did but did it in my head). Putting numbers to it is nice though because I often try to shoot for no more than about 6% crystal and this shows me that I could increase that by 25% and it shouldn't be overly sweet.
Thanks!

I like this idea and the comparison of the three recipes. There are so many wonderful recipes for a pale ale out there, this site is home to a few, but combining three that you like most is ingenious. I can't wait for the results.

What a great write up. I use this sort of technique all the time, either manually merging recipes or doing it in BeerSmith.

@Nagorg really? So now average is not good enough?
Awesome write up OP!

Beer Nerds Unite!!
This is an awesome experiment. I think anyone who creates their own recipes based off of others goes through a similar process in their head (but weighing some obvious biases towards their preferences). Pretty cool to see it all worked out.
Can't wait to hear how it turns out.

better than a below average beer, says I! I'm following along to see the outcome! Great write up!

@Nagorg It'll produce the nectar of the Gods! I shall call it...YoopWortMuncher!
@BadWolfBrewing I plan on translating this technique to as many styles as possible, and hopefully following up with the results.
Thanks to everyone for all the awesome comments!

Cool experiment and interesting approach but I'm not really sure about its viability. You have to consider every malt in a recipe not just for what it intrinsically contributes but also its relationship to other malts.
What might be missing from this approach is incidence of malts. You've essentially done what Ray Daniels did in Designing Great Beers. Look at the "recipes" in that book and they give the percentage usage of malts as well as their incidence. He doesn't suggest to use every malt listed, but rather gives the data so you can decide which are the most important.
But anywho, I'm interested in your results!

@kanzimonson Duely noted! I found it interesting because I took three proven recipes, found relations between the three and developed a recipe. I'm not 100% sure on whether it will work or not, but experimentation advances the practice.

Great idea! Looking forward to the results

Gotta love the DIY ingenuity of homebrewers. Nicely done.

I like that the test batch is 1 gallon. Very nice work.

While I know this is far from the same thing last summer when making bbq pulled pork, i realized i didn't have enough bbq sauce. I decided to mix the last quarter of 6? different bbq sauces figuring what's the worst that could happen?... Turns out that it was the best bbq sauce i have ever made... and i can't duplicate it. If this turns out well at least it could be duplicated. I believe it has promise! will be interested in trying if it turns out!

Well, you know there is always one in the crowd... I couldn't resist the average comment! LOL
This was an interesting article and perspective on recipe formulation.
@Titan88: Looking forward on to the tasting notes for "YoopWortMuncher"!

Being an engineer by degree, this part of brewing fascinates me as well. I love the numbers and the calculation and seeing it all turn into a tasty beverage in the end. Sort of a celebration for completing all the math. This is quite an interesting approach and I am very curious to see how it ends up and how it applies across styles of beer. Bravo for coming up with these experiments. This is what home brewing is all about!

My only thought is; in for a penny, in for a pound. You did the work, the beer is based off three legendary pale ales; why not make five gallons?
Nice work, by the way!

You might want to post the 5-gallon recipe. My brain can't comprehend such tiny amounts of grains and hops.

Very nice wright up.
Cant wait to hear how it taste.
Cheers

Nice break down. I agree though with BadWolfBrewing. The acid levels of the hops vary much more than just going by ounces.

@landshark While that is perfectly true, the hops graph was more about where each recipe added hops in. I could still take the average IBU of each recipe and choose a hop based on distribution and average IBU. Also, if you have higher AA hops, you can just slide them down the scale to a lower time addition, but you can't do that with grains. So, I felt it was more important to have a concrete grain bill, and just get an average distribution of the hops to work with. Great point though!

Here's a 5 gallon recipe that follows your numbers fairly closely, but I did switch the hops around for target IBUs and ease of measuring.
"Average Pale Ale"
Date: 1/27/2014
Author: GlenF
Size: 5.0gal
Efficiency: 75.0%
Attenuation: 75.0%
Calories: 172.45kcal per 12.0fl oz
Original Gravity: 1.052
Terminal Gravity: 1.013
Color: 10.8
Alcohol: 5.1%
Bitterness: 43.3
Ingredients:
10.0oz (6.4%)American Caramel 20L - added during mash
3.0oz (1.9%)American Caramel 60L - added during mash
If it were me, I'd change the 60 minute hop variety and add dry hops, but this should get people started.

I was far less scientific about it, but my usual pale ale recipe (aside from the various SMaSHes that I do) is a blend of BierMuncher's and Yooper's recipes...

I'd like to thank everyone for the comments. I decided to round off my recipe numbers and stick with a one gallon batch because I'm making a cross-country move in less than a month.
OG: 1.056
FG: 1.016
ABV: 5.26%
IBU: 39
SRM: 8
Eff: 70%
Grain Bill:
1.25# Two-Row
0.50# German Vienna
0.25# German Munich Light
2oz Crystal 20
1oz Crystal 60
Hop Schedule:
.5oz Cascade @ Flameout (I had an oz total so I just threw the rest in)

So I did a little better on efficiency than in previous batches. The OG came out to be 1.060, or 75% efficiency. After two and a half days, fermentation has ceased at 1.014. It'll be a touch over 6% ABV. I'm going to give it some more time to condition before bottling. I'll write Part 2 then.

@Titan88
Thanks for the update. Looking forward to the results and part 2.

I'm looking forward to a similar condensation of other styles!

Wondering how this turned out?

I really like this approach..
- Anyone brew something close to this?
Did the OP ever get on to do more of these? (I'm feeling lacy!)

Great post! You can call it the Large Law of Pale Ale's haha Ad infinitum you end up a the perfect recipe.
On a similar note, I wonder if blending yeast strains will yield in better attenuation, timing, etc. ? The short-comings of one strain can be supported by the strength of another.

@Titan88
Did you by any chance ever get around to do one for IPAs?