Not All Thanksgiving Leftovers Are Created Equal


  [Please consider supporting Food and Farm Discussion Lab with an  ongoing contribution of $1, $2, $3, $5 or $10 a month on Patreon. All contributors receive a subscription to our email newsletter the FAFDL Dispatch. Or you can make a one time donation via PayPal. ]

I spent this Thanksgiving with a friend’s family. A fine day and a wonderful evening. As we were pushing the last bits of stuffing around our plates and starting to contemplate pie, my host and I got chatting about leftovers. We had a big crowd and a lot of food and she was balking a bit at the task of leftover management.

We ended up having an interesting conversation about the perverse incentives and cognitive biases that lead to excessive Thanksgiving leftovers. I think the amount of unnecessary leftovers can be reduced by making these incentives and biases explicit and planning accordingly.

No One Likes Running Out of What They Want

What immediately sprang to mind, for me, were the problems of estimating how much you need of each dish not to run out and the asymmetric risks of running out compared to the risks of not having enough. During a brief stint as the executive chef at a retirement community, I developed a keen sensitivity to the incentives to generate excess leftovers when cooking in batches for large groups (as a opposed to the individual choice and preparation in restaurant cooking – or even a simple family meal).

Here’s the problem. We had to feed 160 people for dinner every day, providing a choice of two entrees – one fancy, one more casual. Most days, it wasn’t too hard to guesstimate how many portions of each entree. 90 to 100 people would get the fancy choice and 60 to 70 would get the casual option. Running out of either was particularly infuriating for the residents, because it was the people who had been waiting longest to place their order that also were told they they couldn’t have what they wanted. They really hated it when we ran out. So to be safe, we prepared 105 of the first choice and 75 of the second choice or 180 portions, 20 more portions than diners. I was sensitive to this because I was on a budget, so over-shooting had repercussions for me, but not quite as painful as running out.

It got even more unpredictable and wasteful – that is harder to manage – when we had a really popular dish like steak (a rare treat). Almost everyone would almost always get the big favorites. That meant preparing up to 160 portions of first choice, because running out wasn’t just an inconvenience for our customers, it was a tragedy. But sometimes it would break funny, so we needed 40 to 50 portions of the second choice. The irony being that a predictably popular dish that should have brought our waste down to near zero instead resulted in overshooting by 50 servings instead of 20 (don’t worry, very little was wasted, it all went back into soups and casseroles, but it still is diminished value from getting it right the first time).

Planning Thanksgiving for a crowd compounds these problems five fold. In our case we had we had 10 dishes instead of two. And a smaller group means greater unpredictability – a few outliers can create much larger variation when expressed as a percentage, and that uncertainty is multiplied across a greater number of choices. So the incentives push you towards massive and undifferentiated production overruns.

Lets look at the math

18 diners, 10 dishes, projecting 1.5 servings of each dish = 270 servings for 18 people or 15 servings person. 27 individual portions of each dish. A formula for waste.

(I chose 1.5 servings of each dish as an assumption because I think that’s how most people roughly estimate how much they need – some people will have one serving, some people will have two – shoot for one and half each. )

As my host and I tried to think through how one could address this, we realized a number of things that you need to come to grips with if you want to avoid excessive Thanksgiving leftovers.

Not All Leftovers Are Created Equal, Plan Accordingly

We made a few key observations. The costs of running out versus having too much are asymmetrical, but not consistent across dishes. The value of various dishes as leftovers is not consistent across dishes. There are some dishes that we want leftovers, therefore leftovers do not become waste, but this doesn’t hold true for all dishes.

1. The default assumption that it is better to have too much than too little, is typically applied equally across all dishes. It shouldn’t be. While running out of turkey would be unforgivable, running out of cranberry sauce before everyone has gotten seconds is probably OK. While people might freak out if you didn’t have cranberry sauce on the table, no one will really care if you run out.

2. There are some dishes that you can safely assume will not be eaten by everyone, thus not only did we not need 27 portions of green bean casserole for 18 people, we probably didn’t even need 18. 12 would have probably sufficed.

3. Not all leftovers are created equal. The value of four ounces of turkey on the Sunday after Thanksgiving is nearly equal to the value of those four ounces on Thanksgiving. Meanwhile the value of four ounces of roasted Brussels sprouts on Sunday has cratered since Thursday. Production planning needs to take into account future market value of each dish and price that in. The cost of overshooting production of Brussels sprouts is then quite high, while the cost of overshooting production of turkey is quite low, because while the sprouts will take up room in the fridge and then get thrown out sometime next week, the extra turkey is going to wind up in open faced turkey sandwiches on Sunday.

Projected comparative value of leftovers
Projected comparative value of two dishes

4. Planning is complicated by the fact that many side dishes are brought by guests. The host isn’t in control of their guesstimates. This can be somewhat addressed by taking charge of essentials – which is standard anyways, and then asking guests to bring enough for two thirds the number of people you are expecting – as a rule of thumb, obviously depending on the dish you can make estimates of what is needed based on the considerations above. They will almost certainly bring one and half servings for two thirds of your guests, which sounds about right.

5. The cost of running out of pie is very high. The costs of having too much pie are vanishingly low. Everyone will have pie. Pie is great the next day. And the day after that. Leftover pie is great. You can never have too much pie.

  [Please consider supporting Food and Farm Discussion Lab with an  ongoing contribution of $1, $2, $3, $5 or $10 a month on Patreon. All contributors receive a subscription to our email newsletter the FAFDL Dispatch. Or you can make a one time donation via PayPal. ]

Read More:
• Ask a Farmer: Katie Olthoff, Squaw Creek Turkey Farm
Thanksgiving Soup
• FAFDL Thanksgiving Reader 2015


Please consider supporting by ongoing contribution of $1, $2, $3, $5 or $10 a month on Patreon.
Become a patron at Patreon!

1 Trackback / Pingback

  1. Thanksgiving Soup | Food and Farm Discussion Lab

Comments are closed.