Thursday, October 25, 2012

Best-yet Malnati's style deep dish pizza

[Update (8/26/13): Even though the instructions in this post are very good, I think I've learned a lot about deep dish since publishing this post. Consequently, I may write an even better best-yet post pretty soon.]

After baking a deep dish pizza every day for at least a month, making slight changes with every pizza and every batch of dough, I'm confident that I've created a near-clone of Malnati's deep dish. Better yet, I can teach you how to make a near-clone of Malnati's.

Become a fan of this blog on Facebook!
The most useful pizzamaking lesson you'll ever learn.

I've made a few conscious decisions to stray from what I know is "the right way" to make Malnati's style deep dish (like using only corn oil in the dough, instead of using corn oil and olive oil). Still, this stuff is primo, and you will not find another source that can teach you how to make a better Malnati's style deep dish pizza. If you don't believe me, just scroll down and look at the pictures. Then go look at actual Malnati's pictures. (Just realize that not all of those pics are actually of Malnati's pizza. Also, you may notice that real Malnati's pizzas look a little different than the pizzas shown in Malnati's advertising/menu pics, just as my pizzas look a little different than their ad/menu pics. There's a reason for that.)

Removing the first slice from a Malnati's style deep dish pizza that just came out of
the oven. Hopefully you can see the melted mozzarella oozing from the sides of the slice.

Slice of a Malnati's style deep dish pizza.

Here's some good news: This style of pizza is very easy to make and very hard to screw up, but only if you follow the instructions of someone who knows what they're talking about, like me (or Ed). As long as you have a scale and a pan that resembles a deep dish pan, you should be able to make a very good deep dish pizza. And if you don't have a scale, you need to get one because it is impossible to accurately measure ingredients by volume (measuring cups). Furthermore, if you don't have a scale, it is impossible to measure the proper amount of dough to use for any particular size of pizza. I'm here to help you make great pizza that you can reproduce any time you want, not merely edible pizza that will be different every time you make it. It's easier to make good pizza than it is to make bad pizza, but only if you make a choice to forget everything you think you know and, instead, follow these instructions precisely. And if your first pizza doesn't come out as good as you hoped, try again. You'll do better with each try.

The dough formula for this pizza is very simple. As you read the short list of ingredients, you will surely notice that I include neither cornmeal nor salt in this dough. Here's why: Contrary to what you've heard almost everywhere else, Malnati's dough (i.e., real deep dish dough) contains neither cornmeal nor salt. Here's proof.

So anyway, here's my formula (and here's a post that will help you understand what it means). Update (8/20/13): You might want to decrease the hydration by about 5% because my formula creates a very soft dough, which I'm beginning to think may be too soft.

100% Pillsbury bleached all-purpose flour
58% Water
22% Corn oil

To make 18 oz of this dough (which is a little more than enough for two 9" pizzas), use the following recipe:

9.97 oz Pillsbury bleached all-purpose flour
5.78 oz Water
2.19 oz Corn oil

If you need to make a different quantity of dough, or if you have a metric scale, just go to the dough calculator on and plug in the percentages I've listed above, as well as figures that more appropriately suit your needs. This calculator is very easy to use. If you have any trouble understanding how to use it, just mess around with it for a few minutes.

Specialized equipment you'll need for this pizza:
  • A kitchen scale that measures in ounces.
  • A tin-plated steel deep dish pan or an aluminum cake pan. If you don't have either kind of pan, just use whatever you can get your hands on. Here's a post that teaches you how to season a deep dish pan. (Scroll down to below the picture of a pan with tools in it.)
  • Baking stone (optional but recommended).

Here are step-by-step instructions for how to make this dough by hand. I've worked hard to try to make these instructions easy to follow yet informative at the same time. I have included additional information in many of the picture captions, so be sure to read the captions if my instructions aren't clear. Start by gathering your ingredients: Pillsbury bleached all-purpose flour, ACTIVE DRY YEAST, corn oil, and 110-degree water.
  1. Measure the appropriate quantity of yeast (0.5 tsp) and put it in a custard dish or small bowl. Optional but recommended: Add a pinch of sugar to the yeast.

  2. 1/2 teaspoon of yeast with a pinch of sugar.

  3. Measure the appropriate quantity of 110-degree water (5.78 oz) and pour about half an ounce into the custard dish. Stir the yeast water.

  4. Immediately after adding less than an ounce of 110-degree water to the yeast and stirring.
    Notice that you can clearly see the grains of yeast, as they have not been hydrated yet.

  5. Measure the appropriate quantity of flour (9.97 oz) and set aside.
  6. Measure the appropriate quantity of corn oil (2.19 oz) and set it aside. (The yeast jar in the following pic is just there for scale, to give you a good visual idea of how much corn oil to use.)

  7. Showing about 2.2 oz of corn oil and a jar of yeast (for scale).

  8. When the yeast water is foamy on top, as pictured below, move on to the next step. (Even if it's not foamy after hydrating for ten minutes, go ahead and move on to the next step, but be aware that your yeast may be dead.)

  9. Yeast water after hydrating for 10 minutes. You cannot see any grains of yeast, and the top
    is somewhat foamy and bubbly. This is a clear sign that the yeast is 1) fully hydrated,
    and 2) alive. This indicates that it's time to mix the dough.

  10. Add all the liquid ingredients to your mixing bowl (water, yeast water, corn oil).

  11. Water + yeast water + corn oil.

  12. Add the flour to the mixing bowl.

  13. After adding 9.97 oz of all-purpose flour to the wet ingredients.

  14. Stir the mixture until it all comes together into a cohesive mass of dough.

  15. I mixed the Malnati's style deep dish dough entirely with the spoon.
    It just takes a minute or so.

  16. If you have a plastic dough scraper, use it to gather any small pieces of dough that may be sticking to the sides of the bowl.

  17. Even though the Malnati's style deep dish dough mixes up pretty cleanly, I used a plastic
    dough scraper to gather all the little bits of dough that stuck to the bowl.

  18. Place the dough in a sealable plastic bag and refrigerate (unless you plan to make a pizza within the next 3-5 hours). If you plan to make a pizza within the next 3-5 hours, move on to Step 2 of the next section of instructions.

  19. Even though I may have used this Malnati's style deep dish dough later the same day,
    I still bagged it all up and put it in the refrigerator.

OK, so that's how you make deep dish dough by hand. You can use this dough as soon as three hours after you mix it, but if you keep it refrigerated you can use it at any time over the next 48 hours (or possibly longer). The following instructions are useful either if you plan to make the pizza considerably later in the same day or if you plan to make the pizza a day or two later. (If you plan to make the pizza as soon as possible, follow these instructions immediately after you mix the dough, but give the dough at least 3 hours to rise in the pan and make minor intuitive changes when necessary.)

The following instructions are to make exactly one 9" deep dish pizza.
  1. A couple hours before you intend to bake a pizza, remove your bag of dough from the fridge.
  2. Remove the dough from its bag and scale 8.2 oz of dough. (Put the remaining dough back in the fridge.)

  3. About 8.5 oz of Malnati's style deep dish dough, which is a little more
    than my guidelines suggest for a 9" pizza. (It worked just fine.)

  4. Optional: Grease the pan by pouring a little corn oil or pure olive oil into the pan you'll be using. Even though I think Malnati's oils their pans pretty liberally, I prefer to oil the pan very lightly or not at all. The following picture shows a pretty heavily oiled pan.

  5. Oiling the pan before placing the dough inside to warm up or rise. Even though I think
    this quantity of oil is probably pretty representative of how much oil Malnati's uses, it
    is more oil than I intended to use, and it is more oil than I prefer to use. I have used both
    corn oil and pure olive oil to oil the pan, and both work fine.

  6. Place the dough in the pan.

  7. I distribute the oil evenly throughout the pan before placing the dough in the pan.
    This is about 8.2 oz of Malnati's style deep dish dough, and it's in a 9" pan. Usually when I
    put dough in the pan, the dough is cold because it has been in the fridge for a day or two.

  8. Cover the pan. After an hour or two, set your oven to bake at 450 degrees, with a baking stone on the bottom rack.
  9. After the oven has preheated for at least half an hour, uncover your pan. The dough should be noticeably bigger than it was when you put it in the pan, but it shouldn't be anywhere near double the original size.

  10. The same dough as the previous picture, but exactly two hours later. Notice that it has
    clearly risen over the 2-hour warm-up period, but it hasn't even come close to doubling.
    This is when I prefer to use it, but it's OK to give it another two or three hours.
    I'll show a five-hour rise of similar dough toward the end of this post.

  11. Flatten the dough so it occupies the entire bottom of the pan, then crimp the outer edge of the dough so it reaches about 1-1/4" up the side of the pan. (Brush some melted butter onto the dough if you desire. I can't taste any difference when I do this.)

  12. Malnati's style deep dish dough after forming. The sides go up to about 1-1/4".

  13. Place about 6 oz of sliced mozzarella on top of the dough, then add sausage or any other toppings atop the cheese.

  14. Small pieces of raw sausage on top of 6 oz of sliced mozzarella. (Don't be afraid
    to do this. The sausage will cook.) Malnati's probably uses a little more sausage than I did.

  15. Spoon or ladle about 9 oz of chunky canned tomatoes atop the cheese and toppings. (Scroll way down to learn about the tomatoes I think are perfect for this pizza.)

  16. This is about 9 oz of tomato strips on what will soon be a 9" Malnati's style deep dish
    pizza. I use this tomato product straight out of the can. Even though most of my
    other pics show enough tomato to cover the cheese entirely, I like to use a little
    less than that; sometimes even less than what I've used in this pic. If it looks a
    little wet to you, it's because it is a little wet. That's OK. I have also tried using
    these tomatoes with a heavy puree to make it less watery, but it was nowhere
    near as good as using the tomatoes straight out of the can.

  17. Sprinkle romano or parmesan cheese on the tomatoes, as well as some oregano if you like oregano.

  18. After shaking a little parmesan cheese onto the tomatoes.

  19. Place the pan in your 450-degree oven and bake until the outer edges have slightly browned and pulled away from the side of the pan. (This should generally take about 22-25 minutes.)

  20. Baking a Malnati's style deep dish pizza on the bottom rack in the oven,
    on a preheated stone at 450 degrees.

  21. Remove the pizza from the oven and cut it. (You can cut it right in the pan and serve it in the pan, which is standard if you dine in at Malnati's, or you can remove the pizza from the pan and cut it on a peel or cutting board.)

  22. Malnati's style deep dish pizza, immediately after baking for 29 minutes at 425 degrees.
    Showing pan gripper in action. I've come to prefer baking at 450 degrees for 22-25 minutes.

    Malnati's style deep dish pizza with one slice gone.

And here are some other pics of finished deep dish pizzas:

Bottom of the Malnati's style deep dish pizza.

Profile of a Malnati's style deep dish pizza slice.

Whole Malnati's style deep dish pizza.

Profile of a Malnati's style deep dish pizza.

Whole Malnati's style deep dish pizza.

Whole Malnati's style deep dish pizza, with pepperoni under the sauce.

The following information is my suggested dough, cheese, and sauce weights for every size between 6" and 14". The formula I've used to calculate these numbers is not perfect, but I think it's pretty close. If the sides of your pan are sloped, you should use the pan's bottom diameter.

Even though I have six different sized deep dish pans, I've only used the 9" pan since I began my quest to perfect deep dish pizza. This means the ingredient weights I list for a 9" pizza are more trustworthy than every other size. Consequently, I'll list the suggested 9" ingredient weights before every other size.

9" deep dish pizza
8.16 oz of dough
6.00 oz of cheese
9.06 oz of tomatoes

6" deep dish pizza
3.82 oz of dough
2.59 oz of cheese
3.91 oz of tomatoes

7" deep dish pizza
5.09 oz of dough
3.57 oz of cheese
5.39 oz of tomatoes

8" deep dish pizza
6.53 oz of dough
4.71 oz of cheese
7.10 oz of tomatoes

10" deep dish pizza
9.97 oz of dough
7.45 oz of cheese
11.25 oz of tomatoes

11" deep dish pizza
11.97 oz of dough
9.06 oz of cheese
13.68 oz of tomatoes

12" deep dish pizza
14.14 oz of dough
10.82 oz of cheese
16.35 oz of tomatoes

13" deep dish pizza
16.49 oz of dough
12.75 oz of cheese
19.25 oz of tomatoes

14" deep dish pizza
19.02 oz of dough
14.83 oz of cheese
22.39 oz of tomatoes

Attention: My recommended dough weight creates a pizza that's probably too thin for most people's tastes. Having made many more deep dish pizzas since I wrote this post, I've made some changes to the suggested ingredient weights of a 9" pizza, which I've described in the update at the bottom of the post. As you may have guessed, I prefer the updated weights, even though I'm not sure if they accurately reflect the weights of a real Malnati's pizza.

Sauce for deep dish pizza

For deep dish sauce I use Stanislaus Saporito Filetto di Pomodoro (tomato strips, pictured below), uncooked and with nothing added. Sometimes I use the tomatoes straight out of the can, but I'm beginning to prefer a sauce that's slightly less chunky. To make the tomatoes slightly less chunky, I pulse them a couple times in a food processor. Pulsing them just a couple times keeps them chunky but makes the chunks a more appropriate size. It really is that easy.

Unfortunately, you probably will not be able to find a store that stocks these tomatoes. (If you live in central Ohio, you can get them at Carfagna's, on 161 just east of I-71.) If you are unable to find this brand of tomatoes, I suggest buying canned whole tomatoes. You can either coarsely chop them on a cutting board or pulse them about five times in a food processor.

A lot of knowledgeable home pizzamakers use cans of diced tomatoes mixed with cans of crushed tomatoes, but in my opinion diced tomatoes are too firm to make a good deep dish pizza. Also cans of diced tomato contain way too much water. Whole tomatoes are softer and fleshier than diced tomatoes, and they are almost always packed in puree. So not only do I consider them better tomatoes for this kind of pizza, but it also takes a lot less work to prepare them.

Important: Do not cook the tomatoes. If you want the tomatoes to be less watery, just drain the water (or preferably, drain some of the water).

Stanislaus Saporito Filetto di Pomodoro (tomato strips).
If you can find a place that sells this by the can, buy it.
You can get a very similar product here.

Stanislaus Saporito Filetto di Pomodoro tomato strips. This product works very well
for deep dish. You don't have to use this particular tomato product, but this and the
next pic give you a good idea of what kind of tomato product you want
to use for Malnati's style deep dish pizza.

Another way of showing you the texture of these tomatoes.

Updates and Improvements (12/11/12)

I've made some serious improvements since composing the rest of this post, beginning with a slight formula change:

100% AP flour
52% Water
0.5% ADY
22% Corn oil

To make a little more than enough dough for two 9" pizzas (20 oz), here is a dough recipe:

11.46 oz AP flour
5.96 oz Water
0.57 tsp ADY
2.52 oz Corn oil

Here's a procedural change that seems to work much better than what I've instructed above: Once you've measured all the dough ingredients, add the flour to the mixing bowl, then add the corn oil to the flour. Whisk to incorporate the oil uniformly into the flour. Once you've done that, add the water and yeast water, then mix as normal.

IMPORTANT: One thing I cannot emphasize enough: UNDERMIX THE DOUGH!!! Basically just mix the dough until it all comes together. (It will still be very lumpy when it's finished mixing.) Once you reach this point in mixing, if there are still small pieces of dough and oily clumps of flour on the side of the mixing bowl, just scrape them off and add them to the main piece of dough. DO NOT KNEAD THIS DOUGH, as gluten development is a very undesirable characteristic with this style of dough.

Get it?

After much trial and error, I now think it's better to use more dough and cheese per pizza than I've instructed above. For a 9" pizza, here are my current dough and topping weight recommendations:

8.5 to 9 oz of dough
7.5 oz of sliced mozzarella
Raw sausage by feel (optional)
8-10 oz of uncooked, chunky, fleshy tomatoes

Note: 7 oz of cheese seems to be the magic number. That is, if you use more than 7 oz, you'll end up with a gooey, beautiful mess of cheese that oozes off each slice when you remove the first few slices from your cut pizza. But if you use any less than 7 oz, the magic probably won't be there. If you want mega mega gooey cheese, try using 9 oz of mozzarella.

One last thing: I've been baking my most recent deep dish pizzas at 470 for about 17 minutes (rather than 450 or 425 for 22-28 minutes). I think I like the pizzas better this way, but I haven't decided for sure yet.

Update (6/11/13)

Someone on followed these directions and started a thread about it, which includes at least one picture. Even though most of the posts on that thread seem to be about how I'm a total dick, you might still find it somewhat interesting. (I think maybe I'm glad I was banned, because I'm a better person now than I was when I wasted most of my time on that site.)

Also, a reader of this blog (in Japan, where it's very hard to find ingredients for this kind of pizza) sent me a couple pictures of the pizzas he made after following these directions several months ago. I intend to write a new post eventually, highlighting his results (and hopefully other people's results, too), but it might take me a while to find the motivation. If there is anyone else out there following these directions, I'd love for you to share your results, too. Thanks.

Update (7/31/13)

Here's something I just found that mentions me and links to this page. It's from John Dellavechia, the same person who mentioned this post on His pizzas look real good and have much better photographic quality than mine.


  1. Howdy! Let me first state the fact that you sure have customized a beautiful resource. Also I want to know one thing. Have you ever taken place in any kind of competitions among bloggers?

  2. You're a little light on the oil. Uno's/Malnatis is quite greasy, so you need to up the oil by at least a third.

  3. Thanks a lot for sharing on this useful procedure with us for sure your idea is working best for me.

    Pizza In Zion

  4. I can't believe no one has trolled you yet.

  5. Would you say your recipe is the same as Malnati or is the recipe the best approach for the home cook with all the usual home cooking restraints? I am surprised that you use APF and that you under mix to such a degree. Not that I think you aren't right, it just surprises me that you want no (or really little) gluten development. Have you thought of using a lower protein flour? Seems you would have more of a margin of error with a lower protein content but maybe I am not fully understanding the logic behind your process. I wish you showed a photo of a cut crust so we could see the structure.

    1. That's a tough question. I have only had Malnati's once. However, I have had Uno quite a few times, so I know the two are very similar. It's been a long time since I've had Uno, though, and I've heard that the pizza at Uno's chain/corporate stores (which does not include the original Uno and Due in Chicago) is nothing like it used to be. I don't know if they have changed, though, because I haven't been to Uno in over 10 years. (I have never been to the original Uno or Due.)

      After re-reading your comment, I realize your question has a little different meaning than what I interpreted at first. Probably a good answer to your first question is: I try very hard to make this blog useful to beginners and experienced pizzamakers alike, but I refuse to dumb it down. That is, I refuse to give certain instructions (like using volumetric measurements, particularly with flour), even if those instructions might seem easier for the average person. If I share my recipes in volumetric measurements (which I cannot do, because it is impossible), it might be easier for a beginner to understand today, but that same person's pizza will most likely suck, both today and every other time in the future.

      I'm not here to give bad instructions. Rather, I'm here to help people make several different styles of pizza almost exactly how they make them where you buy them. This blog exists because I know most of the secrets. I have no interest in steering anyone in the wrong direction.

      It might take a whole 30 minutes to read about bakers' percentages and fully understand it, but those 30 minutes may cut out hours of work from every pizza you make in the future. Also, your pizzas will be infinitely more consistent, and you will be able to scale your dough batch to any size, which you can't do with volumetric measurements. It's not harder to do it this way vs. the way we've all been taught. In fact, it's easier. Much easier. Infinitely easier.

      I cannot tell you how many cups of flour to use for any particular recipe, nor can anyone else, because there is no such thing as one cup of flour. Every cup of flour has a different weight, even though every level cup of flour looks the same. 3 cups of flour might weigh 12 oz or it might weigh 18 oz. You just never know, because your technique for measuring flour volumetrically might be the total opposite of mine, which is why you really should not measure flour volumetrically. If you use measuring cups to measure flour, the amount of flour you use from one batch to the next will never be the same, which is very imprecise, thus pointless.

      There are some ingredients you'll never be able to measure by weight in a home environment--like salt, sugar, and yeast--unless you buy an expensive scale that can accurately measure fractions of a gram. That's not a big deal, though. It's OK to use measuring spoons for these ingredients, because if your measurement is off a little (with these ingredients), it won't change things much (or at all).

    2. One thing I think is very obvious with real deep dish pizza (like Malnati's or Uno) is that there is very little gluten development. To me their crust is very much like a biscuit: flaky, crumbly, etc. If you develop the gluten very much, you cannot have that flaky, crumbly characteristic. "Gluten development" is a term you hear a lot nowadays, which you didn't hear much prior to about 5 or 10 years ago. Everyone seems to love saying "gluten development," but most of the people who say that have no clue what it means. And really, I'm not even sure you can know what it means until you've played with dough about 1,000 times. Maybe the term 'gluten development' is useful in other areas of baking, but it is not very useful in pizzamaking.

      You know how sometimes you sink your teeth into a pizza, or more likely, bread, and then your jaws hurt by the time you're finished chewing that bite? That's gluten development. That's not a characteristic I want in a pizza, regardless of whether I bought it or made it myself. That's why I don't use All Trumps flour anymore. Full gluten development might work with bread, but it doesn't work with pizza. Despite the fact that pizza and bread are both categories of baking, that is about the only similarity.

      Pizza is not bread; especially deep dish pizza. But essentially no style of pizza is bread. If you use breadmaking strategies to make pizza, you'll end up with pizza that seems like bread, which is not a good description of any pizza you've ever bought.

      Lower protein flour? I actually did consider using a lower protein flour for a different style of pizza not long ago, but I eventually decided AP flour is about the lowest-protein flour I would want to use for any kind of pizza. For almost all other styles, I will not use flour from a grocery store. Similarly, with deep dish, I would like to try a foodservice-quality flour with a similar protein percentage. Still, I am very OK with using Pillsbury or Gold Medal AP flour for deep dish.

      And I do really like those two flour brands; I don't like anything about King Arthur flour. I don't know why so many people think King Arthur flour is so great, but to me it seems like most of those people are just victims of marketing. Believing King Arthur makes great flour is like believing McDonald's makes great hamburgers or that Budweiser makes great beer.

      Honestly, even though I think my Malnati's style deep dish is every bit as good as Malnati's, I have a ton to learn about deep dish, and I know I'm not doing everything how Malnati's does it. Part of that is because not only do I want to clone their pizza, but I also want to make it better. I am inclined to think Malnati's probably mixes their dough for longer than I do, and with a different attachment (dough hook), but I really like how it comes out when I mix it either by hand or with a flat beater (for much less than a minute), which is why I do it that way.

      Sorry it took so long for me to reply. I haven't really paid any attention to this blog in several months, and I didn't even see your comments until today. It's not that I don't want to pay attention and maintain the blog. It's just that things are different than they used to be, and I can no longer do a lot of things I used to be able to do. The fact that I have even written this comment is a big-time win. I haven't even checked my email since last November.

      From what I remember, it seems you may have left a couple other comments elsewhere. If you leave any more comments, I probably won't see them for a long time (partly because I don't ever check my email anymore). So if you do leave any more comments, you might also want to text me to let me know, because I do check my texts. (SixOneFour) SevenThreeEight-ThreeEightSixSeven.