Math formula for Weight Watchers’ points

Factual Questions
1

I have joined Weight Watchers, and am getting accustomed to their system of “points.” Every food is assigned a point value and, depending on your weight, goal, and several other factors, you are allowed a certain number of points per day. They have a really great web site with all sorts of helpful tools, including a data base with values for a huge number of foods. And if a given food is not listed, there’s a little box in which you can type in the data (Fat, Carbs, Fiber, Protein), and it figures out the number of points. I’m trying to figure out what formula they use to arrive at the number of points. Here are some values for a selection of foods:

http://www.flickr.com/photos/42009607@N00/5752669429/

Given the data for Fat, Carbs, Fiber and Protein, what is the formula for determining the number of points?

2

Are you sure that chart is current? Under the new plan (which was launched in early December 2010, IIRC), most fruits and vegetables are “free” (0 points), excluding obvious exceptions such as peas, corn, potatoes, and avocadoes.

I can say that the new points calculation, whatever it is, is more complicated than the old one. With the old plan, you could roughly guess that 50 calories was about 1 point, adjusting up and down depending on fat (more, fat, more points) and fiber (more fiber, fewer points). Not so with the new system. I kind of fudge it sometimes, going by the old system and figuring that the new one dings you more for carbs.

The other thing is that with the old system, you got a free paper slider in your tracker that you could use to figure points, because there were only 3 parameters: calories, fat, and fiber. With teh new system, the formula has four parameters and is apparently too complex for a slider. So everyone has to buy a calculator (or use the online one).

Good luck figuring out the formula. I’m not sure it can be done by working backward from points charts. I’m also pretty sure that Weight Watchers considers the formula proprietary information and guards it pretty closely. On the message boards they don’t even like people to give points for non-WW recipes.

(If you want to be a real stickler, you would refer to “POINTS values” for the old system and “PointsPlus” for the new. As if everyone doesn’t just call them “points.”)

3

Yeah, that chart looks wrong. The banana, blueberries and tomato should all be 0 points. A can of Coke (12 oz) is 4 points. Some of the values are correct, such as the egg and the milks. A can of tuna (6 oz) should be 4 or 5 points for a can, not just 1 point.

Where did this chart come from?

4

I’m aware of that. I’m deliberately not using the “Points Plus” values, but am trying to derive points from the raw data. All I’m looking for is a mathematical formula, disregarding the “PointsPlus” system.

5

On the “Plan Manager” page, there’s a small area on the left that enables you to input the 4 values for any food, and a “calculate” button to give you the number of points. I have done this for every food on my list. Again, I’m only looking for the math formula, not something that corresponds to Weight Watcher’s system.

6

completely ignorant and uninformed, but did WW points simply used to be kilo-calories divided by 100 and rounded up to the nearest integer?

I thought that’s what it was back in the day when I was still living at home and my mom was on the program (despite being thin and obsessively fit.)

7

Now I’m really confused. You want to come up with the formula for Weight Watchers points, but you don’t want the result to correspond to WW points? At first I thought you were just trying to figure it out as a mathematical exercise (similar to how people are always trying to figure out the Coke formula and whatnot), but if you don’t want the answer to match the WW points, then I’m not sure what the point (heh) is.

And if you’ve joined Weight Watchers, one assumes you want to follow their system. So why not just use the points values as given? (Why “pay” 3 points for a banana when the WW system says they’re “free”?) True, sometimes there are minor discrepancies between the online database, the WW books, and the calculator, but usually not as great as some of the values you’ve given.

Take tuna. The can of tuna in my cupboard (Chicken of the Sea, solid white albacore tuna in water, 5 oz) has 1 g fat and 13 g protein, fat and carbs 0 g (per serving, 2 servings in a can). Your chart lists 0.5 g fat (perhaps you have a different brand, no big whoop), but the rest is the same.

My online calculator gives 1 point per serving; doubling the values gives 3 points for the whole can. (Another thing to keep in mind: because of rounding, 1 + 1 sometimes equals 3. On the old plan, Progresso light clam chowder was 1 point per 1/2-can serving, but 4 points for the whole can. Caveat eater.)

The online database lists 4 points for 5 oz of canned chunk white tuna in water.

My pocket WW calculator gives 1 point per serving (1/2 can) and 3 points for the whole can.

My tracking booklet lists 3 points for 4 oz of canned tuna in water, drained.

So I’m curious as to how you got your online calculator to arrive at 1 point for an entire can of tuna, unless you neglected to allow for serving size. That’s very important.

I guess it would help me out if you could explain your ultimate goal.

8

This would be a lot easier to figure out with simpler data: Put in values for hypothetical foods that are pure fat, pure carbs, or pure protein. Also, because they round the final number of points, we could get more precision by using very large portion sizes. But let me see what I can do with the data we have.

Point values are just added up, right, and you’re supposed to stay under some total per day/week/whatever? And if the plan makes sense, then it should give the same total number of points for two foods separately as for a dish made of those two combined. From that, the formula (neglecting fiber, which I’ve a hunch will be treated separately) should be something of the form afats + bcarbs + c*protein. There are seven foods on that list without fiber; that should be more than enough to figure out a, b, and c.

Let’s start with the simplest food on the list: The can of Coke, with pure carbs. The formula gives 3 points for 27 carbs (I presume that’s grams?), meaning that the value of b should be about 0.11 (or at least, between .093 and .13). Next we have the olive oil, with pure fat. 14 fat gives 4 points, giving an a of approximately .29 (between .25 and .32). Finally, we need a value for c, the protein coefficient. There isn’t any food on the list that’s pure protein. The tuna’s close, but at only 1 point, we’re going to have significant roundoff error, so let’s look at the chicken instead. The 7.6 fat in the chicken should contribute 2.17 points, leaving 2.83 points from the 29.2 protein. That gives us a c of approximately .097 (between .070 and .12).

OK, now, if we did this right, we should be able to use these values to find the number of points for any of the other non-fiber foods on the list. Let’s try the egg. It has 5.0 fat, .6 carbs, and 6.3 protein, so it should have .295.0 + .11.6 + .0976.3 , for a total of 2.13 . And lo and behold, it’s listed as being 2 points. How about the whole milk? That’s 8.0 fats, 12.0 carbs, and 8.0 protein, so it should have .298.0 + .1112.0 + .0978.0 , or 4.416 . That’s a bit off from the 4.0 they list, but still within roundoff error.

Really, one should enter all of those data points into a single calculation and solve for all of the parameters at once, but I’m satisfied that it’s basically right.

9

He (panache is a he, right?) wants the formula that’s used on the Weight Watchers’ web site, where you can enter those four numbers for a food and get the total number of points. The values from that formula might differ from the tabulated values for specific foods, but he’s just curious about the formula.

10

According to Wikipedia…

points = max(0, round(protein/10.9375 + carbs/9.2105 + fat/3.8889 - fiber/12.5))

Link

11

…And actually, it occurs to me that there might be a simpler way to get this. If you go to the page with the formula, and view the source code, the calculation is probably just done in Javascript, and you can just read off the exact numbers they use. It’s not like they’re any great secret; anyone could have done the same sort of analysis I just did, so there’s not much point in them trying to hide the numbers.

EDIT: Or you could just find it on Wikipedia. That works, too.

12

Thanks, guys, my question is answered. I should have just given you the math, without even mentioning Weight Watchers. :smiley:

13

The girlfriend is on the new plan and it is a total WTF?! Kidney beans are high in points relative to their nutritional value. Apples are 0 points (should be about 2 points) with the same amount of sugar as a flavored water which is points.

14

Dunno about the beans, but the apples undoubtedly get significant benefit from their fiber content. Ain’t no fiber in your flavored water, but whole apples have lots.

15

I’m not currently on WW so I don’t know about the new plan, but I understand why apples are 0 whereas the flavored water has points.

Not only is it the fiber, as has been mentioned, but it’s also to incent people to make healthier eating decisions. I recall that many people would choose to eat a 1 Point snack bar rather than a 2 Point banana because if you looked at it from a pure Points perspective the processed snack was a better choice (but only because it cost you fewer points, not because it was healthier). This way fruit is a better ‘value’ Points wise, so people are not going to be put off eating it for that reason.

Personally I always thought of fruit as a freebie. I know it wasn’t the proper way to deal with it, but I lost 50 lbs so it worked for me. The only way I could stay on track was to have a go to list of foods I could eat with abandon when I had a serious jones to snack, so I allowed myself as much raw veggies and fruit as I wanted. YMMV.

16

WW made some choices with the new plan that relate to common habits more than just nutrition. Having most fruits and veggies be 0 points is a simplification that encourages participants to eat more of what is generally good for you and something that people don’t overdose on anyway. They had a similar scheme in the old points plan where you eat an infinite amount of certain foods and only take a set number of points. That addresses how people actually eat rather than the exact calories/fat/fiber.

17

Mods:

Please close this thread. I intended it to be merely a math problem (which has been solved), not a debate about Weight Watchers. Thank you.

18

Before the thread gets closed, a small correction: Not an INFINITE amount of certain foods (generally most fruits and veggies, whole grains, low-fat dairy, and lean protein), but rather you can eat them until you’re satisfied. You use your weekly points for anything that isn’t on that list of foods. It’s called the Simply Filling method and it’s still in effect. I’ve never tried it.

19

And using the standard figures of 4 Calories per gram of carbs or protein and 9 Calories per gram of fat, this works out to 1 point per 44 Calories for protein, 1 point per 37 Calories for carbs, or 1 point per 35 Calories for fat. So the plan favors carbs over fat, but only a little, and protein over either.