As noted in my earlier post, I've been losing weight this year.

Results

In particular, to give some numbers here, this is how my weight has changed over the course of this year:

• 16/01/11 - 117.1 kg
• 13/02/11 - 113 kg
• 13/03/11 - 110.9 kg
• 17/04/11 - 109 kg
• 15/05/11 - 107.4 kg
• 12/06/11 - 105.7 kg
• 17/07/11 - 101.2 kg
• 14/08/11 - 98.9 kg
• 18/09/11 - 99.3 kg
• 16/10/11 - 98.2 kg
• 13/11/11 - 95.6 kg
• 11/12/11 - 93.9 kg

So, clearly, it's working :-)

Method

In summary I've done this by:

• Keeping a detailed food diary - using rednotebook. I've used it to count calories. The purpose behind this was self-education about how many calories different foods really contain.
• Calculating my RDA once per week using this formula
  Calories = ( 10 * weight in kg + 6.25 * height in cm - 5 * age + 5 ) * activity factor
  (The formula for women is - (10 * weight in kg + 6.25 * height in cm- 5 * age - 161) * activity factor )
  Note this takes into account activity levels. Fat people (like me) tend to have a very low activity factor. ("I have a low metabolism" is usually asserted without it being tested, and often wrong).
• Aiming to eat about 75% of RDA per day, at most, but not worried too much if I go over that once or twice a week, and taken care to go over 85% as little as possible and not go under 60% as much as possible.
• Measuring the weights of food I have no idea of their weights to calculate RDA, used smaller plates, and bowls that hold known, more sensible, portion sizes.
• Snacking primarily on protein rich/calcium rich/low fat foods to control hunger/snacking urges. (babybel light primarily)

All of this is explained in detail in the earlier post.

Crucially, two things I have not done:

• I have not particularly changed my exercise levels. I walk to/from the station, and cycle occasionally in good weather in summer. I do have much more energy now though. Or rather I probably have the same amount of energy, but more is available for things other than shifting my weight around.
• I have also not denied myself foods. Yes, I've had cake, pizza, chips etc. I've just learnt what a sensible portion size looks like.

Observations

Based on this, I've got sufficient data - 47 weeks' data - to create a heatmap / scattergraph based on:

• Amount of calories "over" eaten per week relative to RDA - -25% means 75% of calories eaten. I've picked "over eaten", because it puts RDA down the middle of the graph. These are not daily averages, these are relative to the weekly calorie allowance.
• Amount of weight gained

Which enables me to examine the reasonable hypothesis - does eating more that RDA correlate with weight gain, and under eating correlate with weight loss ?

Well, the graph below is based on those 47 weeks' data.

To me, what that says is pretty clearly - yes, over eating relative to RDA does correlate with weight gain, and undereating RDA does tend to correlate with weight loss.

Looking closer, if you over eat over RDA, there are 6 data points all showng weight gain. If you look at the 85% point, there are 5 datapoints which relate to weight loss, and 7 that relate to weight gain, and between 85% and 100% being 1 with gain and 5 with loss. Below 85% there is 1 datapoint with weight gain and 34 with weight gain.

Some tentative conclusions

Whilst correlation does not imply causation, let's presume it's a reasonable conclusion.

If that's the case, this means we can derive some guiding principles from this, ordered by how useful I think they are:

• Weeks where you eat consistently less than 85% of RDA will result in weightloss.
• If you're feasting (eg christmas, easter), then feast, then stop. The worst thing you can do is (for example) feast solidly for 2-3 weeks in a row. The alternative is to replace the usual foods you would eat with the christmassy foods if you just want to eat christmassy foods over an extended period.
• It's the long term (week/fortnight) RDA calorie count that matters.
• Short term (2-3 day) calorie restriction (eg "detox diets") or calorie binges (eg family birthday party weekend) may have a short term effect, but no long term positive/negative effect.
• There's no apparent benefit of going below 70% of RDA.
• Weeks exceeding total RDA for the week will result in weight gain.
• Working on a weekly calorie budget is easier/less restrictive than a daily denial based allowance.
• Weeks with weight gain, will gain about a pound or two (0.5-1kg)
• Weeks with weight loss, will lose about a pound or two (0.5-1kg, average 0.7kg)
• If you've piled on weight, day in day out, for 2 years, it will take you 2 years to lose it.

I'm not claiming that these conclusions apply to others, but based on the numbers, I do think that they apply to me.

Slightly Silly, But Nice Observation

There's one final point, according to RDA I "should" have eaten 792,235 calories. I've actually eaten 638,000 calories (about 80% overall), and lost 23.2kg. That's 154235 calories, or 6648 calories per kilo. That's about the same amount of calories as in a kilo of sunflower margarine. There's makes a rather nice parallel.

So, under eat by a tub of margarine, and lose a tub of margarine-worth of weight. Very literal, and very true.

What do you call this diet then

I've been asked by a few people what diet I'm doing. I think I call it the "putting less in one end than comes out the other diet". Bit gross but pretty accurate :-)

That or just "I'm doing the eating less / not stuffing my face diet".

Does also mean I can just enjoy food, as long as I make it count. But that's pretty fun, 'cos food is fun :-)

For those that know me, most of you unlikely to have ever known me not to be fat - even though I grew up chronically underweight. If you've known me for a while, you've probably seen me try and lose weight, and may've seen me when I lost around 10kg back near the start of 2007. Whilst that was the lightest I'd been in probably 12 years even then, that weight came back on with a vengence afterwards though. This was probably due to using a "don't eat this, that, other diet". The problem there is it's natural to stop. 2007 was a hard year (as was 2008/2009) for personal reasons, and a denial diet simply doesn't work then.

Anyway, for the past few years I'd tried making a new years' resolution to try and be lighter at the end of the year, than at the start of the year, and to try and lose weight each month. That resolution has each year failed.

This year I resolved not to make any new years' resolutions. However, around a week into the year, BBC 1 (or 2) repeated a show called "10 things you need to know about losing weight". Unlike many recent BBC science programmes, rather than banging on 1 point for an hour, it spent 5 or 6 minutes on each of the 10 points, running mini-experiments on each. Each to illustrate a point. Back at the time I was impressed at the level of content and how much they backed up the content that I sky+'d it and made some detailed notes. (I'm basing this post on those notes from then and also thoughts since)

The short summary of the 10 points is as follows:

1. Don't skip meals - Skipping meals makes your hungrier, but also means the brain responds more (measured using an MRI scanner) to higher calorie foods - meaning when you eat next, you eat more than you would otherwise. Demonstrated with people painting a bridge.
2. Change your plate size from 12 inches to 10 inches - Studies with popcorn packets found people finished around the same proportion of their popcorn, whatever size the popcorn packet was. Simply using a smaller plate leads can lead to eating 22% less and not noticing. People with more food available eat more.
3. Count your calories - Demonstrated that most people can't guess what food has most calories using - Two sandwiches; Muffin; Few small squares of chocolate; Large plate of potatoes, lots of greens, chicken breast.
4. Don't blame your metabolism - Monitored what a fat actress was eating by using a camera/video camera and food diary. On the surface she was eating sensibly, but when added up was eating around 1000-1500 more calories per day than she should. She also had her metabolism measured (which she blamed), and in fact her metabolism was normal - she was just eating too much.
5. Protein staves off hunger pangs - Showed what happens when the stomach shrinks as it empties of food. Chemical called ghrelin gets released - to tell brain stomach is empty. Eating protein appears to block reception (or reduce production) of ghrelin, meaning you don't feel (as) hungry.
6. Thick soup keeps you feeling fuller for longer - They made a chicken, rice and veg meal for two groups. Both were served with a glass of water - one was blended with the food to form a thick soup. Both groups had their stomachs sizes scanned using ultrasound. Those with the glass of water had the water absorbed the food quicker - meaning their stomach shrunk back in an hour or two - than those with the water blended in. That means the soup people felt fuller for longer.
7. The wider your choice, the more you eat - This was a fun demo - two bowls of smarties, place somewhere saying "eat me, free". One has a single colour of smarties, the other has lots of colors. The multicolour ones go first. The implication here is at a buffet you're more likely to go back because of the "have you tried the ..." effect.
8. Low Fat Dairy helps you excrete more fat - Some research shows that calcium levels in dairy bind well with the fat in dairy, meaning it's more likely to pass through you "all the way" implying low fat dairy has "spare" calcium. Demonstrated this by feeding a volunteer two known diets for a week, one week high dairy, both with same fat content. Week with high dairy led to higher fat in faeces, meaning unlike other week it hadn't been absorbed.
9. With exercise, the afterburn is what matters - Measured metabolism immediately after exercise and 10-12 hours later. Found that the effect of exercises isn't necessarily how long you exercise for, but whether it's sufficiently long enough. Essentially, if you use up your carbohydrate reserves, the body "burns" fat until it's had a chance to rebuild its carbohydrate reserves again. Point being that a bit regularly was better than rarely and lots.
10. Keep moving to lose weight - Possibly the most lightweight part of the programme. Made the point that it's possible to integrate a bit more exercise if you look at journeys to/from work and how you work.

As I say, each point was backed up with the help of tools to measure metabolism and plausible demonstrations/recreations of various pieces of research.

So, for me the key things were:

• Know how many calories there are in things, don't base portions on crockery size.
• Low fat dairy (or stuff with calcium in) is useful.
• Don't get over hungry - manage your ghrelin levels using soup, protein and eating regularly.
• You need to exercise enough, and with some thought can build it into your day. How much exercise is "enough" depends on the person.

This really boils down to - eat a variety, eat low fat dairy, count calories, don't over eat, and build some exercise into your day.

New Year's Resolution

As a result, I resolved to start counting calories - NOT to change what I was eating.

My reasoning here was simple - if I had no idea what I was eating in terms of calories, how would I know when I'd reduced them? The other reason is simple - it's not a denial based thing - it's something that allows me to keep on doing it whether or not I'm under or over eating.

Around the same time I'd been looking around for something to use for taking notes, and came across rednotebook. Rednotebook is organised like a diary, in that there's a page per day, and uses a very simple markup for writing notes. It provides simple tag and word clouds, along with a simple full text search. I've been using it all year.

Counting, Measuring and Monitoring

Anyhow, based on that I started writing down everything I was eating or drinking. For things involving ingredients I've weighed things in advance, and then found a container that holds that much and used that as a measure since. After a little while I realised that, like many people, I fall into food habits. This naturally leads to it being quicker to write things down and their calories.

So, have I measured every day all year ? No. There have been 2 periods this year when I've not done that. The first was near easter when one of our cats was run over. It hit us as a family hard, and I really didn't want to do anything. Since it was easter, I estimated those two weeks intake in terms of calories based on previous weeks. The latter was around my birthday, and it's totally impractical to count every sweet - so I took the tins of sweets, found out the calories in them and divided them out over the days I ate them - again as a realistic guestimate.

The final step though is that I'm a geek. I figured if I'm going to be OCD about it in order to learn how many calories there are in different foods and learn better habits, the more I monitor things the better. As a result, yes, I've dumped it all a spreadsheet and been using that to monitor things. On the one hand, I felt a bit dirty for using a spreadsheet - after all, shouldn't I be using a decent programming language, but on the other I've gained an insight into why some people use spreadsheets as much as they do.

So, to give you an idea of what this gives me - over the past 284 days I've eaten 551,332 calories, or an average of 1941 calories per day. For comparison, if I was eating the recommended daily allowance (RDA) for calories I "should" have eaten 688,708 calories, or an average of 2425 calories per day. That means I've "undereaten" by 137376 calories. Given I've lost about 21kg of weight, that means for each 3000 calories I've undereaten I've lost a pound of weight.

Anyone who's looked at lots of diets will "know" that this matches roughly what people say, but it's interesting (to me) to see it stand out so starkly.

So, why have I stuck with this? Have my habits changed ?

Changing Habits Requires Understanding

I think the reason why I've lost weight is for a number of reasons, from previous diets:

• Whilst it's possible to eat 3000-4000 calories in a single sitting, you don't automatically gain a pound of weight.
• Doing something like Atkins shows you there's a "burn in" period for diets to work.
• Doing a little exercise for a day or two has less effect than regular exercise.
• Measuring weight daily is pretty pointless, since changes you measure relate to the weight of the food you ate. (Takes food up to 48 hours to go through the digestive tract)
• So, working on a daily basis for intake is probably a bit silly too.
• Eating too little generally results in piling on weight afterwards

So, I decided to find aformula for calculating RDA, and use that to calculate a weekly figure. There's a few out there, but the one I'm using is this:

Calories = ( 10 * weight in kg + 6.25 * height in cm - 5 * age + 5 ) * activity factor

Where activity factor is:

1.200 = sedentary (little or no exercise) This is the figure I pick
1.375 = lightly active (light exercise/sports 1-3 days/week, approx. 590 Cal/day)
1.550 = moderately active (moderate exercise/sports 3-5 days/week, approx. 870 Cal/day)
1.725 = very active (hard exercise/sports 6-7 days a week, approx. 1150 Cal/day)
1.900 = extra active (very hard exercise/sports and physical job, approx. 1580 Cal/day)
(The formula for women is - (10 * weight in kg + 6.25 * height in cm- 5 * age - 161) * activity factor )

So, in practice I do this:

• Calculate an RDA figure as above, once a week.
• Set a "target" usage of 75%. Going over this is fine, going under is fine, though if at the end of the day it's below 60% eating some toast or cheese or whatever to bring it up to a minimum of 60% . This target will probably shift to 85% as I get closer to my target weight.
• To treat this like a budget and average things over a 7 day period.
• This means some foods are more expensive (in terms of calories) than others. If go over budget one day, going under on another makes up for it. If you go over all week, with money you'd need an overdraft. With calorific overspend, you get an over draught - as you belly over hangs your belt and gets a draft.
• No foods are off limits - remember my resolution was to count calories, not restrict them.

This combination does mean that I have to count everything, and that's a bind. However because no foods are off limits, I can make a guess of "is this OK", followed by later estimating the actual amounts. The upshot of this is immensely freeing.

Snacks

I found in early days that I snacked a lot, and that whilst individual meals were generally OK, some of the snacks weren't. Some examples of snacks (and what they're equivalent to)

• A few biscuits and a chocolate can easily hit 400 calories (vs a good lunch at Nando's)
• A venti latte with shot of syrup, and a sticky bun - 800 calories (vs a deep pan cook at home pizza)
• Flapjack - 400 calories (vs see above)
• Packet of Doritos - 1300 - (vs a large deep pan cook at home pizza)

Clearly I wouldn't eat all those things often, but the point being is that I'd snack differently on different days, but ultimately, I'd snack. In part this is because I'd skip breakfast, or skip lunch. One of the killers here though is that many many "cheap" (or snack) foods are high in carbohydrate, sugar and fat, but low in protein. The sugar and fat make them more-esh, the carbs make you think they're filling, but the effect wears off, and the lack of protein means that they don't really satiate you. The upshot is high (but empty) calorie intake which just makes you gain weight.

Changing Behaviours

What I did is I decided to work with the way I ate rather than against, though in a more informed way.

• I found out the volume that cereals of different types came up, because buying a bowl that was approximately the recommended portion size. (A "normal" cereal bowl - eg the sort given away for free - is about 4-6 portions(!)). Since then for each new cereal, I've weighed the volume of cereal that fills the bowl. This means that I never need to weigh that cereal again, and can just calculate based on that.
  • The sort of bowls I'm using for cereal are glass and about 8cm across. This sounds small, but you rapidly get used to it, and seems about right.
  • This means I get breakfast every morning, for between 230 - 300 calories depending on what I have.
• As a result, when I get hungry at lunchtime, I go for lunch. If it's the canteen, I'll have the soup which contains meat (for the protein). If I go to the Lowry Mall, I'll get a baked potato with cottage cheese (for the same reason). Soup + bread is around 250-300, potato +cottage cheese is about 350.
  • Before we moved to MCUK, I'd get a "lighter choices" ready meal from Tesco express and cook it in the microwave at work, since that'd be about 400 calories.
• This leaves me then with an average of 1000-1200 calories for evening meal, which is more than enough for a good meal.

So, for me snacking was caused by two things - habit and hunger. The latter is managed by:

• Regular meals
• Eating 1 or 2 babybel light cheeses when I'm hungry and a meal is "too far off" - low fat, high protein - to suppress the ghrelin effect.
• Actively choosing snacks which will reduce the hunger effect.

Regarding reducing habits, I've dealt with that by buying puffed wheat (like sugar puffs which haven't had anything added - no sugar, syrup etc). I've then tried flavouring that with things like peri-peri salt, or with sucralose type addition. That gives something bulky to snack on. If it's late at night, I'll either have a babybel or toast, etc. Also, if I really feel like a sweet snack, I'll have something like a Cadbury's brunch bar - because it's a got clear indication of calorie cost, but it's also not so small you think "I'll have another". Simply being aware of eating has made me snack less.

Clearly I still go out to the shops, but if I have a coffee whilst I'm out, rather than having a latte, I'll have an Americano. Similarly, if I'm having something to eat there, instead of a cake, I'll either get a small pack of mini muffins or wafers to share, or a cheese toastie. The former would be for snacky habit reasons, the latter is if I'm peckish/hungry. (And a cheese or ham/cheese toastie again is protein rich)

Similarly, there's been other revelations - for example Nando's can be a remarkably healthy lunch or meal out as a treat - having rice instead of chips, having a side salad, and similar can mean that it's lower calorie than a "meal deal" of sandwiches, crisps and coke from boots.

Exercise?

From the way which I'm monitoring things, it's clear that exercise helps. However it rains here a lot and I'm not going to go to the gym. Ever. As a result, the vast bulk of my weightloss hasn't had anything to do with extra exercise. That said, now that I'm over 20kg lighter, I do unsurprisingly find climbing the stairs to the 5th floor at work less of a struggle, cycling is easier, and walking from Old Trafford to work is quicker than getting the tram, so on nice days (or days where I must be there by a certain time), I'll walk from Old Trafford. (though there is a remarkably tempting fryup/cafe on that route..)

However, it's a misnomer IMO to say that you must exercise to lose weight. What would be worse is to start exercising, and not change your diet. Primarily because when you stop exercising your diet will still be just as bad, and you'll put all the weight back on. Given when I was between the ages of 11 and 14 I'd be cycling 20 miles a week, rising to 36 miles a week between 15 and 16, and then to about 115 miles/week between 16 and 19.

Oddly enough, continuing to eat the same sorts of food I'd grown up to eat between 11 and 19, when cycling, dozens of miles a week (or day) results in significant weight gain with a sedentary lifestyle. It's just a pity that it took me so long to realise that the habits I gained growing up didn't really fit my adult lifestyle. (I was chronically underweight as a teenager/young adult - probably due to the amount of cycling)

So, whilst exercise is important, understanding how you're eating weight is more important, and especially the relationship between exercise and diet.

I'll note this though - cycling to work for a month before going on holiday meant that the decreased weight loss over the course of the holiday was offset by the increased weight loss when cycling. So being a fair weather cyclist is better than not cycling at all. Similarly, measure how long you can walk and how far.

When I started this, my limit for walking quickly was chunks of around 20-25 minutes. That hasn't really changed (I am still overweight by about 20kg after all). What has changed as I've lost weight is that my speed has improved. What used to take 20,25 minutes now take about 11 or 12 minutes.

Also, my BMI was 34. Someone thin with a BMI of 21 or 22 saying "hey just exercise more" hasn't got a clue. Start off doing something of 10-20 minutes that leaves you out of breath. That might be walking to the shops or walking to the station. Or walking to a different stop. But do it. It'll help start a virtuous circle. Try to do something every day if you can - including the weekend - even if it's just window shopping.

What else?

So clearly this approach is working. I note everything down, find out figures for foods from things like supermarket websites, from web searches, from packets etc. I put it all in a spreadsheet and use that to track a number of different things. In particular, I have 4 sheets:

• Daily - into this I put the total calorie intake, distance walked, distance cycled, and monitor that relative to 60/75/80/100% RDA, rolling 7 day average, rolling 7 day average of averages.
• Weekly - weight, weight delta, BMI, calories eaten, RDA calories, average weight loss (to estimate dates when will have lost weight)
• RDA Calcs - wraps up the formula above, and various notes/versions to make understanding better
• Monthly - RDA calories vs actual calories vs weight vs weight loss per month.

Some Findings

For these, as sad as it may sound, I've also been graphing these to see what trends emerge.

• It's clear that exercise does increase the rate of weightloss if your calorific intake stays constant
• Eating more calories than RDA over the course of a week causes weight gain, of an average of 0.7kg or about 1.5lb. There's no particular correlation of "amount of calories over eaten" vs weight gain. It seems that there's a limit as to how much weight the body can put on at any point in time.
• Eating fewer calories than RDA over the course of a week causes weight loss, of an average of 0.7kg or about 1.5lb. Again, there's no particuar correlation of "amount of calories undereaten" vs weight loss. Again, it seems that there's a limit to how much weight the body will lose at any point in time.
• The thing of 3000 calories per pound of weight loss does seem to be true. Interestingly, that's also about the same as the amount of calories in a pound of margarine (eg Stork). Worth pondering.

What that means of course is that if you're going to over-eat for whatever reason - be it celebrating someone's birthday, christmas, etc the best approach is to actually feast - ie if you're going to go for a big meal, go for a BIG meal. If you're given a tin of sweets for a birthday, don't try and spin it out, binge on them. You'll have more fun (the reason the person gave you the sweets), and by eating them quicker it'll have less of a long term impact.

Likewise, this also means that if you want to lose weight, you actually have to do it for at least 3-4 weeks to lose any real weight, and even then only expect to lose just under 1/2 stone (nearly 3kg/6lb). Anything else is probably just emptying of your digestive tract and water loss.

The other slightly depressing observation though is this - it'll take at least as long to get the weight off as it took to put the weight on in the first place. I've been doing this "diet" now for 284 days. For me to reach my target weight will be another 280 or so days. Thankfully, the "diet" I'm on allows me to have things like pizza, pasta, cake, cheese, meat, bacon, sausages, cheesecake, cakes, desserts, chocolate, mousse, yoghurt, custard, vegetables, chicken, salad, ice cream, hot dogs, and all sorts of similar foods.

Bottom line

If you want to lose weight count calories (for at least 6-8 months until you get a good feel for calorific cost), aim for 75% of RDA, and don't worry about occasionally going over, or under, unless you drop below 60%, snack on protein if you must (of ideally less than 80 calories), buy some smaller crockery to control portion sizes. (Really hard to understate how much of a difference that makes)

Closing thought

My BMI is now 28. This means I've lost 6 BMI points. My target is the middle of the healthy BMI range 22 - which means I have another 6 BMI points and another 9 months to go. I suspect the next 9 months will be harder than the first 9 months. If you're struggling to lose weight though, losing 3 1/3rd stone, 46lb or 21kg is possible, and doable without a denial diet.

Just focus eating sensibly, as much exercise as you're comfortable with and on the next 1kg or 1lb at a time. That's what I've done. Seems to work.