The Fresh Loaf

News & Information for Amateur Bakers and Artisan Bread Enthusiasts

Characteristics of bread flour?

nicodvb's picture

Characteristics of bread flour?


implementing some of Hamelman's recipes in "Bread" I had to make a lot of adaptions.

Let me explain: the flours I have at disposal in Italy are much less absorbent and elastic than american and canadian flours (that most probably were used by the author for his recipes), as I already imagined beforehand.

Fortunately for me the recipes more or less explain how the dough should look and feel like, permitting me to add  flour to have a dough with seemingly the same characteristics.

My curiosity is understanding what are the alveographic characteristics of the "bread flours" you generaly use. At page 372 there's an example of a flour in use: W 335 with P/L 0.75.

Here that flour would be considered high gluten flour with a very high elasticity. Just to make an example "manitoba" flours (specialty high gluten flours to make sweet breads such as panettone and pandoro) have W ~= 350 and P/L 0.6 with an absorbence of at most 60%. Flours with a higher P/L are generally considered to be too elastic, especially for sweet breads. Ordinary flours in our stores almost never drink more than 50-55% when preparing stiff doughs.

Another example: preferments for rye breads are generally called for with an hydratation of 80-83%, but the rye flours I use are so finelly milled (and wholemeal) that with so little water I would get a very stiff dough that I could only "knead" by hand.


I'd like to have some more detail on your bread flours to have a wider picture.


LindyD's picture

Tucked away at page 145 of Bread is a subtopic labeled "flour choice," wherein Mr. Hamelman makes a recommendation about the "bread" flour to be used, noting it should be of medium strength, about 11.5 to 12 percent protein, with an ash content of about .5 percent.  

The KAF bread flour is stronger than that, coming in at about 12.7 percent protein.  The KAF all purpose flour meets those specs and it is actually AP flour that Mr. Hamelman uses, under the King Arthur professional label of Sir Galahad flour.

I'm not sure if this is the information you were interested in, but the specifications of the various professional flours available through KAF can be read at

Franko's picture


I found this excerpt from Cavel's 'Taste of Bread' on Google Books. It has 3 alveographs , 1 of French Flour, 1 of North American/Japanese and 1 of Canadian flour. The side by side comparisons may be useful to you.


Chuck's picture

It may be most informative to just get a "feel" for the 'manitoba' flour that's available to you. Or you may find this presentation of flour rough equivalencies to be useful. Or to get a flavor of which specs are available and useful in the U.S., see these specs that classify some of the flours some U.S. bakers use. (Note well though that these particular flours are intended for professionals; they are not available in retail stores and are not typically used by home bakers.)

Home bakers often just go by the (non-standardized) marketing terms, such as "bread flour". (Generally the marketing term "bread flour" really means little more than "highest gluten content in this brand".) Home bakers often also pay attention to the protein content and perhaps even the ash content. Even these two numbers are surprisingly difficult to obtain (they're not printed anywhere on flour sacks), but have become a little more commonplace as web technology has spread. Generally "protein content" and "gluten content" are used interchangeably, even though that isn't quite accurate.

(Of course some of the few folks who either mill their own or buy very locally get more numerically detailed than this:-)

Note that these two numbers can be (and usually are) held almost constant for a particular brand and kind. So for example King Arthur Flour "all purpose flour" was 11.7% protein last year, is 11.7% protein this year, and will be 11.7% protein next year. Keeping the gluten content constant like this is typically done by mixing different wheats right at the mill, so the retail buyer is never even aware of any of the details.

Professional bakers also sometimes have access to numbers that vary for different batches of flour, in particular the "falling number" and a summary of the Brabender farinograph results. The complete graphical farinograph is for some reason (too difficult? historical tradition? handy summary? not requiring graphics?) typically distilled down to just four numbers: i) either 'absorption %' or 'arrival time', ii) 'peak time', iii) 'stability' or 'departure time', and iv) 'mixing tolerance index' [MTI]. These number are virtually never available to home bakers.

One suspects (I don't really know for sure) that the flour mills have additional numeric tests which they use internally to more accurately characterize all the various flours they produce, thus being a little more specific about contractually required specifications. (Typically one mill produces flours for several different brands, using the wildly different specifications set by each of those brands. Thus it's not true that all flours produced by a particular mill are of similar quality.) Assuming the mills do have additional tests, the results -in fact the very existence of such tests too- are never made public.

One thing you may notice is there's nothing that even tries to numerically describe the fineness of the grind. Even with something like rye flour, U.S. bakers are generally reduced to having nothing more than the very subjective marketing terms "fine" and "medium" and "coarse". There are no rough rules of thumb associating for example particular gluten contents with particular grind finenesses.

If you want a more detailed mumeric answer, it probably just isn't going to happen. There are a couple huge reasons: First, the tests are different! As you may have noticed reading through the above, tests like "W" and "P/L" just plain aren't done in the U.S. at all ("damaged starch" is sometimes tested, but only rarely). One could argue that the U.S. numeric characterization of flours is just plain inadequate; that experience has shown it to be "good enough" for typical flours, but it has obvious inadequacies when attempting to characterize some specialty (for example some organic) flours. (Then again, one could argue that flour is so complicated that trying to reduce its characterization to a reasonably short string of numbers will always fail in edge cases no matter what.) And second, even the numeric tests we have in common are done assuming different conditions! U.S. measurements are made with 14% moisture content in the flour, whereas my understanding is that the same measurements made in Europe assume 0% moisture content. So even the gluten percentage number needs to be mathematically converted (not as drastic as Fahrenheit to Celcius, but still...).


nicodvb's picture

that the level of details I'm searching doesn't seem to be available to the public, that is a real pity.

Yes, numbers alone don't tell a lot, but they are a good starting point. For example, if I know that a certain flour won't stand a too long mixing time Ill try to exceed overly long mixing; on the other side if I know that a certain flour won't rise all that much or that it's too elastic I won't use it to make a panettone that must grow 4-5 times in volume. Looking at synthetic numbers I can make at least a first good selection of what flour NOT to use.

Just to make a couple of concrete examples: in order to make bread I'm currently using an average T55 flour with a nominal protein percentage of 13%. Judging from that value one could believe that it's a very high gluten flour with high absorbence, instead it's quite poor in both respects and it doesn't even stand long mixing, so if I had used it for my pandoro I would have made a very hazarded choice.

For my pandoro (50% of butter, 31% of sugar, 4 yolks) I chose a much stronger flour that on the envelope reads just a meager 11% proteins. It took me 50  minutes of working, but the dough came together perfectly and with an optimal consistence.

In this case the protein percentage would have been a very misleading indicator.


American and European flours should really join forces and come out with a unified, thorough and useful benchmark system. The current state of business is very sad!


Thanks everybody for your answers.

JoeVa's picture

Ciao Nico. Here what I know.

Avarage W and P/L for American All-Purpose flour are available from Heartland Mill. In the "Quality and Technical Information" web page (here the LINK) they publish lab analysis for each lot number. Just look at the AP flour, they have the unenriched and the enriched version. The last one includes malt to adjust the falling number. In short this flour has:

- W: 230-330
- P/L: 0.8-1.4
- Protein: 10.5-11.5 (US)  ->  12.5-13.5 (EUR)

Moreover remember that European protein content is expressed as a percentage of dry matter. In the United States and Canada it is calculated on a basis of 14% flour humidity. So United States 11% = Italian 13.1% protein content.

Maybe it's a great flour, but only from these alveograph analysis I can say this is a very inconsistent unbalanced flour. There's a huge "fluctuation" in the W and P/L is very high (super-long autolyse needed)!

Just compare this with our millers. Even the small one where I buy my flour (Marino) is much more consistent, and it use organic grain too!

- W: 250-290
- P/L: 0.45-0.60
- Protein: 11-11.8 (US)   ->   13-14 (EUR)

You know these numbers are not directly applied to the buratto flour you get in the bag (type 1, T80) but to the same grain mix milled in the lab to a more refined flour. So the higher extraction flour you get out from the stone grinder is a bit different ... it soak up lot of water, up to 75-80% for a medium-soft dough!

In conclusion AP is not so different from our good quality medium strength bread flour (W 280).

But I have to say one more thing about the hydration point. Just think at the same flour splitted in two bags, BAG10 with 10% humidity and BAG15 with 15%.

If you make a test dough with 60% hydration on the flour from the bag you have:

BAG10: (90g flour + 10g water/humidity) + 60g water

BAG15: (85g flour + 15g water/humidity) + 60g water

Total water content of BAG10 test is 77% on the dry flour

Total water content of BAG15 test is 88% on the dry flour

To get from BAG15 the same total water content of BAG10 we have to use 51g of water. From 60% to 51% hydration.

... I don't know if this make sense!!


nicodvb's picture

with you Giovanni, a range as wide as 230-330 or 0.8-1.4 is so large that it's essentially a joke: it doesn't make any sense.

I know that Marino flours are excellent and very absorbent, sooner or later I'll buy something from them.