Senin, 25 November 2013

Dried mussels: A little plate with 160 g of protein (plus some comments on high-protein low-carbohydrate dieting)


Many hunter-gatherer groups employed various methods of drying to preserve meats. Drying also increases significantly the protein content of meats; this is the case with dried mussels. I discussed this effect of drying before here with respect to small fish (). The photo below is of a plate with about 240 g of dried mussels that I prepared using the simple recipe below.



To prepare your mussels as in the photo above, you will have to steam and then dry them. You can season the mussels after you steam them, but I rarely season mine. Almost none of the food I eat requires much seasoning anyway, because I use nature’s super-spice, which makes everything that has a high nutrient content taste delicious: hunger ().

- Steam the mussels for about 10 minutes, or until all are open.
- Remove the mussels from the shells; carefully, to avoid small shell pieces from coming off into the mussels (they are not kind to your teeth).
- Preheat the oven to about 200 degrees Fahrenheit, and place the mussels in it (on a tray) for about 1 hour.
- Leave the mussels in the oven until they are cold, this will dry them further.

About 240 g of mussels, after drying, will yield a meal with a bit more than 160 g of protein – i.e., the proportion of protein will go from about 20 percent up to about 67 percent. In this case, most of the calories in the meal will come from the protein, if you had nothing else with it, adding up to less than 800 calories.

This comes in handy if you need to have lunch out, as the dried mussels can be carried in a plastic bag or container and eaten cold or after a light re-heating in a microwave. To me, they taste very good either way; but then again anything that is nutritious tends to taste very good when you are hungry, and I rarely have breakfast. I often eat them with pre-cooked sweet potato, which I eat with the skin (it tastes like candy).

You may want to think of dried mussels prepared in this way as a protein supplement, but a very nutritious one. You will be getting a large dose of omega-3 fats (3.11 g) with less omega-6 fats than you usually get through fish oil softgels (where n-6s are added for stability), about 1,224 percent of the recommended daily value (RDV) of magnesium, 461 percent of the RDV of selenium, 1,440 of the RDV of vitamin B12, a large dose of zinc, and (interestingly) almost 100 percent of the RDV of vitamin C.

Since mussels are very low in the food chain, accumulation of compounds that can be toxic to humans is not amplified by biomagnification (). But, still, mussels can be significantly affected by contaminants (e.g., petroleum hydrocarbons), so sourcing is important. The supermarket chain I use here in Texas, HEB, claims to do very careful sourcing. Telltale signs of contamination are developmental problems such as thin shells that shatter easily and stunted growth ().

For those readers who are on a low-carbohydrate diet, please pay attention to this: there is NO WAY your body will turn protein into fat if you are on a low-carbohydrate diet, unless you have a serious metabolic disorder (see this post: , and this podcast: ). And I mean SERIOUS; probably way beyond prediabetes. Do not believe the nonsense that has been circulating in some areas of the blogosphere lately.

A high-protein low-carbohydrate diet is one of the most effective diets at reducing body fat, particularly if you do resistance exercise (and you do not have to do it like a bodybuilder). That is not to say that a high-fat low-protein diet (like the "optimal diet") is a bad idea; in fact, the optimal diet is a good option if you do not do resistance exercise, but that is a topic for a different post.

Senin, 11 November 2013

Latitude and cancer rates in US states: Aaron Blaisdell’s intuition confirmed


In the comments section of my previous post on cancer rates in the US states () my friend Aaron Blaisdell noted that: …comparing states that are roughly comparable in terms of number of seniors per 1000 individuals, latitude appears to have the largest effect on rates of cancer.

Good point, so I collected data on the latitudes of US states, built a more complex model (with several multivariate controls), and analyzed it with WarpPLS 4.0 ().

The coefficient of association for the effect of latitude on cancer rates (path coefficient) turned out to be 0.35. Its P value was lower than 0.001, meaning that the probability that this is a false positive is less than a tenth of a percent, or that we can be 99.9 percent confident that this is not a false positive.

This was calculated controlling for the: (a) proportion of seniors in the population (population age); (b) proportion of obese individuals in the population (obesity rates); and (c) the possible moderating effect of latitude on the effect of population age on cancer rates. The graph below shows this multivariate-adjusted association.



What is cool about a multivariate analysis is that you can control for certain effects. For example, since we are controlling for proportion of seniors in the population (population age), the fact that we have a state with a very low proportion of seniors (Alaska) does not tilt the effect toward that outlier as much as it would if we had not controlled for the proportion of seniors. This is a mathematical property that is difficult to grasp, but that makes multivariate adjustment such a powerful technique.

I should note that the 99.9 percent confidence mentioned above refers to the coefficient of association. That is, we are quite confident that the coefficient of association is not zero; that is it. The P value does not support the hypothesized direction of causality (latitude -> cancer) or exclude the possibility of a major confounder causing the effect.

Nonetheless, among the newest features of WarpPLS 4.0 (still a beta version) are several causality assessment coefficients: path-correlation signs, R-squared contributions, path-correlation ratios, path-correlation differences, Warp2 bivariate causal direction ratios, Warp2 bivariate causal direction differences, Warp3 bivariate causal direction ratios, and Warp3 bivariate causal direction differences. Without going into a lot of technical detail, which you can get from the User Manual () without even having to install the software, I can tell you that all of these causality assessment coefficients support the hypothesized direction of causality.

Also, while we cannot exclude the possibility of a major confounder causing the effect, we included two possible confounders in the analysis and controlled for their effects. They were the proportion of seniors in the population (population age) and the proportion of obese individuals in the population (obesity rates).

Having said all of the above, I should also say that the effect is similar in magnitude to the effect of population age on cancer rates, which I discussed in the previous post linked above. That is, it is not the type of effect that would be clearly noticeable in a person’s normal life.

Sunlight exposure? Maybe.

We do know that our body naturally produces as much as 10,000 IU of vitamin D based on a few minutes of sun exposure when the sun is high (). Getting that much vitamin D from dietary sources is very difficult, even after “fortification”.

Senin, 28 Oktober 2013

Aging and cancer: The importance of taking a hard look at the numbers


The table below is from a study by Hayat and colleagues (). It illustrates one common trend regarding cancer – it increases dramatically in incidence among those who are older. With some exceptions, such as Hodgkin's lymphoma, there is a significant increase in risk particularly after 50 years of age.



So I decided to get state data from the US Census web site (), on the percentage of seniors (age 65 or older) by state and cancer diagnoses per 1,000 people. I was able to get some recent data, for 2011.

I analyzed the data with WarpPLS (version 4.0 has been just released: ), generating the types of coefficients that would normally be reported by researchers who wanted to make an effect appear very strong.

In this case, the effect would be essentially of population aging on cancer incidence (assessed indirectly), summarized in the graph below. The graph was generated by WarpPLS. The scales are standardized, and so are the coefficients of association in the two segments shown. As you can see, the coefficients of association increase as we move along the horizontal scale, because this is a nonlinear relationship. The overall coefficient of association, which is a weighted average of the two betas shown, is 0.84. The probability that this is a false positive is less than 1 percent.



A beta coefficient of 0.84 essentially means that a 1 standard deviation variation in the percentage of seniors in a state is associated with an overall 84 percent increase in cancer diagnoses, taking the standardized unit of the number of cancer diagnoses as the baseline. This sounds very strong and would usually be presented as an enormous effect. Since the standard deviation for the percentage of seniors in various states is 1.67, one could say that for each 1.67 increment in the percentage of seniors in a state the number of cancer diagnoses goes up by 84 percent.

Effects expressed in percentages can sometimes give a very misleading picture. For example, let us consider an increase in mortality due to a disease from 1 to 2 cases for each 1 million people. This essentially is a 100 percent increase! Moreover, the closer the baseline is from zero, the more impressive the effect becomes, since the percentage increase is calculated by dividing the increment by the baseline number. As the baseline number approaches zero, the percentage increase from the baseline approaches infinity.

Now let us take a look at the graph below, also generated by WarpPLS. Here the scales are unstandardized, which means that they refer to the original measures in their respective original scales. (Standardization makes the variables dimensionless, which is sometimes useful when the original measurement scales are not comparable – e.g., dollars vs. meters.) As you can see here, the number of cancer diagnoses per 1,000 people goes from a low of 3.74 in Utah to a high of 6.64 in Maine.



One may be tempted to explain the increase in cancer diagnoses that we see on this graph based on various factors (e.g., lifestyle), but the percentage of seniors in a state seems like a very good and reasonable predictor. You may say: This is very depressing. You may be even more depressed if I tell you that controlling for state obesity rates does not change this picture at all.

But look at what these numbers really mean. What we see here is an increase in cancer diagnoses per 1,000 people of less than 3. In other words, there is a minute increase of less than 3 diagnoses for each group of 1,000 people considered. It certainly feels terrible if you are one of the 3 diagnosed, but it is still a minute increase.

Also note that one of the scales, for diagnoses, refers to increments of 1 in 1,000; while the other, for seniors, refers to increments of 1 in 100. This leads to an interesting effect. If you move from Alaska to Florida you will see a significant increase in the number of seniors around, as the difference in the percentage of seniors between these two states is about 10. However, the difference in the number of cancer diagnoses will not be even close to the difference in the presence of seniors.

The situation above is very common in medical research. An effect that is fundamentally tiny is stated in such a way that the general public has the impression that the effect is enormous. Often the reason is not to promote a drug, but to attract media attention to a research group or organization.

When you look at the actual numbers, the magnitude of the effect is such that it would go unnoticed in real life. By real life I mean: John, since we moved from Alaska to Maine I have been seeing a lot more people of my age being diagnosed with cancer. An effect of the order of 3 in 1,000 would not normally be noticed in real life by someone whose immediate circle of regular acquaintances included fewer than 333 people (about 1,000 divided by 3).

But thanks to Facebook, things are changing … to be fair, the traditional news media (particularly television) tends to increase perceived effects a lot more than social media, often in a very stressful way.

Senin, 30 September 2013

The sub-2 hour marathon? Don't hold your breath, just yet

Is the sub-2 hour marathon imminent?  Don't hold your breath

Yesterday Wilson Kipsang took 15 seconds off the marathon world record, running 2:03:23.  It triggered, as it always does, talk of how close they are to breaking the two-hour barrier.  But that's very, very premature.  For reasons of physiology, performance evolution, and the inter-connectedness of performances from 10km to the marathon, we are a long, long way from going under two hours.

It's not the same as for you or I, who find ourselves a few minutes outside a barrier, and know that six months of hard training and a good day will break it.  This is a world where the margins are tiny - that's why we can look at the pacing strategy and the splits and comment that perhaps it was a little too fast in sections, when in reality, "too fast" means 1 second per kilometer, accumulated over 20 minutes!  The precision of the physiology to run a 2:03 is extra-ordinary.

So consider for instance the progression.  In the modern era, catalyzed by da Costa's breaking of Dinsamo's 1988 record, the improvements in the record are as follows:
23 seconds, 4 seconds, 43 seconds, 29 seconds, 27 seconds, 21 seconds, 15 seconds.

This record is not going to be "smashed" by anyone.  Anything greater than 20 seconds is a big improvement.

The real story is not how often the record is broken, it's how often it isn't

What the sequence above doesn't say, which is more important, is that for every successful attempt, there are probably fifty (a hundred?) unsuccessful ones, where the best runners in the world are on course for the time, for some of the race, then fall away.  Every year, five or six big city marathons start with high hopes - London, Dubai, Chicago, Rotterdam, Frankfurt, Berlin and perhaps two or three others.  Across these races, there are likely twenty "viable candidates", and yet perhaps one in a hundred will come off, despite intent and incentive.

That's why when you look at the record books, you'll see that there are now about 50 performances under 2:06.  Most of those started out as record attempts, and many will have had high hopes up to halfway, or even 30km.  In London earlier this year, about six of the best marathon runners in history went to 25km on course for the world record.  The explosion was huge, and some ended up finishing outside 2:09, or not at all.  I recall Emmanuel Mutai closing at 5:00/km.  The same happens every year in many races.  Even in Berlin yesterday, only one man of a group of three sub-2:06 guys held on to run a 61:51 second half.  Incredible running, but it should highlight just how rare successful attempts are.

The implications - many factors have to align

In the future, that will become more the case.  As this record drops, it will become more difficult to break, and that has some implications.

First, it requires an absolutely perfect day.  London, where the final 8km are run along the embankment, often finds a headwind that could easily cost 2 sec/km and that would be enough to eliminate record possibilities.  Chicago has found itself too hot or too cold.  Berlin was too wet recently.  Dubai hot or windy.  Unless the weather is close to perfect, the record is becoming too strong to break.

Second, the marathon course must be perfect.  I think there are probably only four or five courses in the world that are viable for a world record.  Dubai, Berlin (obviously), London (though the wind, and the number of turns, makes me wonder whether this is still the case, actually), Chicago, and then one or two of the second tier races like Frankfurt where Kipsang ran 2:03:42 a few years ago.

The marathon course is nothing without the best athletes, and so now you also need huge money to attract the best men, in the right numbers, for a record.  London has in the past been so strong that the athletes watch one another rather than risk losing to chase times.  New York gets amazing fields, but the course compromises the final time.  Second-tier marathons with perfect profiles can't get the depth of quality to deliver the record.

The problem then is that there are only a few opportunities a year for the top guys to have a realistic shot.  Now the above three factors need to come together - you need perfect weather on the perfect course, with the best athlete in close to perfect condition, and suddenly you can see why unsuccessful attempts outnumber successful ones so convincingly.

So, what does this mean?  It means that if the record is broken by 15 seconds each time (I think this is a realistic expectation, particularly as it gets stronger), then one can expect it to happen perhaps once every three years.  More likely four or five in the future, but if it were three, then in order to cut another 3:23 off in 15 second intervals, you're looking at around 40 years.

The physiology and performance links behind the 2-hour marathon

So this talk of a sub-2 hour marathon is so premature.  There are a few physiological reasons why it is also not feasible at this stage.  I have written on this extensively before:
But to sum it up as briefly as possible, the point is this.  If you want a guy to run sub-2 for a marathon, then you're asking for a capability of back-to-back half marathons in under 60 min.  The current WR for the half is 58:23, by Tadese (who hasn't turned that into a decent marathon yet), but for the most part, the top men are running in the low-59s.  The very best break the 59-min barrier.

In other words, the currently best runners on the planet are hovering around 59-minutes for half the distance that people expect them to run in a marathon, at the same pace.  It's a little like expecting Usain Bolt, with his 19.19s 200m best, to go out and run a 400m, slow down just a little, and run a 41s World Record.

Or it's expecting David Rudisha, who can run a 400m in 45s, to hold a pace of 46s for two laps and run 1:32, rather than his 1:41 for 800m.  It just isn't going to happen, and the reason is that the pace we can run for a given distance decreases in a predictable, physiologically 'constrained' manner as the distance increases.

So a man who runs a 59-min half marathon will not be able to sustain two back-to-back 60 min half marathons.  It's just not possible.  And so therefore, before we can even consider the sub-2 hour marathon, we need to look at the ability over the half marathon.  Until humans can run a half-marathon in under 58-minutes (and here, I'm talking low-57), it will not be possible to produce 59:59 twice in a marathon.

And that can be taken one step further, to 10km.  If you are going to see a 57:x half marathon, then you should also be seeing a 10km that is substantially faster than the current 26:x.  The 10km performance required to run a 57 is probably in the high 25s.

It's possible, of course, that the change could come from the ability to sustain high speeds, rather than to nudge the entire system faster.  In other words, the runners of the future could well run at current 21km paces for twice the distance without the paces for shorter distances changing.  That would change the relationship between intensity and duration as we know it, but it is possible if the threshold capacity of runners changes (substantially) in the future.  But that's not going to come instantly - there are physiological barriers that must be inched out of the way, not leapt right over.

Those relate to the physiological implications, which I have written on before, so I won't go into here.

Bottom line is that talking about a sub-2 hour performance after seeing a 2:03:38 improve to a 2:03:23 is just not feasible.  The next barrier is 2:03, and I'm sure will go within five years.  Then we can begin to work towards 2:02, which will take another ten years, perhaps.

It's a great period for marathon running - every season, fall and spring, we get to anticipate a record at least twice.  2013 has delivered a successful attempt, but it shouldn't lull us into expectation that more of the same is just around the corner.

Ross


How to handle a dog attack


For most people, dog attacks are not very common. But they happen occasionally, and the experience can be traumatic. Incidentally, they are also a good reason why I am not a big fan of barefoot walking or running. Broken glass pieces and nails can be a problem if you are barefoot; so can dog attacks.

The photo below, from Dreamstime.com, shows a charging dog. It reminds me of an incident many years ago where a dog attacked my two oldest sons, who were very young at the time. They were unsuspectingly playing at a park in Southern New Jersey, when I saw a dog running in their direction across the park. Part of what I will say in this post is based on experiences like that.

I should also say that I grew up around dogs. My grandfather had a farm that was managed by my uncle, and dogs were critically important in managing the farm. One problem we had was that domesticated pigs would often become feral, or would mate with wild boars, in some cases leading to a particularly vicious breed of large feral pigs. I was once attacked by one of these feral pigs while hunting. One of the farm dogs came to my rescue and probably saved my life.



If you are like most people, when you go walking outdoors, you do not carry a walking stick or a cane. Maybe you should. But if you don’t, thick-soled sneakers can be used in a reasonably effective defense in a dog attack situation.

Dogs attacks’ main targets: The faces of children

Dogs tend to be loyal friends, but they must be monitored for signs of aggression, and can be particularly dangerous to children. A significant proportion of dog attack victims are children 5 years of age or younger, who more often than not sustain injuries to the face, with secondary target areas being the hands and feet ().

At the time of this writing the web sites Documentingreality.com and Arbtalk.co.uk had some grisly photos of dog attack victims (, ). They show evidence that the face is often targeted, and some possible consequences of real dog attacks.

Artificial selection: Dogs and Moby-Dick

Modern dogs are descendants of wolves who came into contact with humans about 12,000 year ago. (This general date is often cited, but is the subject of intense debate, with DNA studies suggesting much earlier contact.) Wolves are apex predators; this was true also for wolves that lived around the time they first came into contact with humans. They hunt and live in packs, and rely on fairly complex body language, a variety of sounds, and a keen sense of smell to communicate.

Even being apex predators, wolves were no match for humans. Therefore, as humans and groups of wolves co-evolved, dogs emerged. Dogs evolved instincts that made them sociable toward and submissive to humans, particularly those humans who fed them and also asserted authority over them – those become their “owners”.

Humans, in turn, came to rely heavily on dogs for protection and hunting, and probably evolved instincts that are still largely unexplored today. For example, there is strong evidence suggesting that having pet animals, many of which are dogs, is generally health-promoting (, ).

The evolution of sociability and submissiveness traits is an example of what is often referred to as “artificial selection”, where animals and plants evolve traits almost exclusively in response to the selection pressure applied by humans. In the case of dogs, this was later taken to new heights through selective breeding; leading to the emergence of a variety of dog breeds, some for utilitarian purposes and others for pure vanity, each with very distinctive characteristics.

Interestingly, artificial selection applied by humans does not always produce more sociable and submissive animals. The opposite happened around the mid 1800s due to excessive hunting of sperm whales. The least aggressive were easier to kill, so they were overhunted. Over generations, this placed selection pressure in favor of the evolution of aggressiveness toward humans. The attack on the Essex by a large bull sperm whale, which served as inspiration for Herman Melville's novel Moby-Dick, was one of the first incidents that resulted from this selection pressure (). Whaling increased, and, predictably, attacks started becoming more and more frequent.

When a dog attacks, stand your ground in a non-threatening way

Dogs, like wolves, are territorial animals. Many dog attacks are likely motivated by humans invading what a dog perceives as its territory at a given point in time. I mentioned earlier in this post that a dog once attacked two of my children. They were playing at a park during the winter. Nobody else was there. I saw this large black dog running from a distance in their direction, and I immediately knew that it was trouble. The dog probably saw us as invading its territory.

Having grown up surrounded by dogs, I pretty much knew what to do. I walked toward my children and placed myself between them and the charging dog. I told the children not to move at all, just freeze. The dog came running until it realized that we were not running. It was a “fake charge”, like most are. It stopped close to me, and barked very aggressively, coming closer. I was wearing boots. I raised one of my boots toward the dog’s snout, and when it bit it, I pushed the boot against its snout.

Here is where I think most people would tend to make a key mistake. They would probably try to hurt the dog to scare it off, by, say, kicking the dog as they would kick a soccer ball. The problem is that, because the dog is a lot faster than they are, if they do that they may end up missing the dog entirely and worse - they may end up losing their balance and falling to the ground. This is when dogs can do the most damage, since they would go for the face of the fallen person.

As a side note, often you hear that dogs attack the throat of their human victims, but that is not what the statistics show. Most victims of dog attacks display injuries on the face and extremities. The "myth" that dogs target the throat is probably based on the notion that dogs attack humans because they see them as prey. However, with exception of feral dogs such as Australian dingos, evidence of dogs preying on humans is very rare. I've reviewed many dog attack photos for this post, and could not find one with evidence that the throat was targeted.

So I pushed my boot against the dog’s snout a few times, firmly but not with the goal of hurting the dog, and did not do anything threatening toward the dog otherwise. This calmed the dog down a bit, but it was still acting aggressively and would not go away. Sometimes firm commands to "seat", "stop", "go away" make the dog react submissively. I tried them but they didn't work; instead they probably made the dog more excited. Then I did what probably is the one thing that most land animals instinctively fear from humans …

Sapiens the thrower

I picked up a few pieces of ice from the ground and threw at the dog. One piece of ice hit the dog on the side of its body; a couple of others were glancing blows. As a result the dog became visibly confused and submissive (telltale sign: tail between the legs), and ran away. Here is where another big mistake may happen. People may try to hurt the dog and become too excited when throwing objects at it. In doing so, they may end up not only missing the dog with the flying objects that they are throwing, but they may also excite the dog, and face another attack.

The best approach here is to focus on having whatever you are throwing at the dog land on top of or as close to the dog as possible; explicitly without trying to hurt it, in part because this improves your aim. Having flying objects coming from you toward the dog is enough to trigger the dog’s instinct to get out of the way of “Sapiens the thrower”. Moreover, if you don’t try to hurt you’ll be relatively calm, displaying the type body language that will trigger submissiveness.

I’ve long suspected that throwing has been a key component of Sapiens’ climb to the top of the food chain, to the point that all land animals have an instinctive fear of humans – even large predators, and much bigger animals such as elephants (as long as they are not “in musth”). One short video has been circulating on YouTube for years; it has various hunting scenes where primitive spears are used (). Many find this video cruel. It clearly shows the enormous evolutionary advantage of humans being able to throw pointy things at other animals. If humans happened to live when Tyrannosaurus rex was around, there is no doubt in my mind that the latter would be the prey.

Keep your face away and your hands closed

Typically you’ll avoid a full-blown dog attack by only standing your ground for a while and not acting aggressively toward the dog. After a short standoff period, you’ll just walk away unharmed. Unfortunately this may not happen if you are facing a dog that has been trained to attack. In this case, having a stick or something like it will help a lot. (In circus acts lions are “pushed around” by trainers holding objects like sticks and wooden chairs; sometimes that doesn't end well - .) If you don’t have one it would be useful to be wearing shoes that can withstand several bites. If not, you can use a piece of clothing, such as a bundled jacket, as a shield.

If you have a stick, or something like a stick, you should not try to hit the dog with it. You should place it near the snout, and push the stick against it each time the dog bites. If you do this calmly and firmly, without trying to hurt the dog (remember, the dog is a lot faster than you are), you will probably discourage biting after a while, turning the attack into a standoff.

What if you don’t have anything with which to defend yourself at first, and a dog attacks you? Keep your hands closed into fists, to avoid having fingers bitten off, and do your best to keep the dog away from your face. As desperate as these situations may be, try to be calm and look for objects that you can use to push the dog away, that you can throw at the dog, or that can be used to wrap around your arms. Frequently there will be objects around that can be of use – e.g., sharp stones, glass bottles, pieces of canvas, loose pieces of a fence, a hose, a tree’s branch. If you fall, try to stand up right away. Very likely you'll sustain injuries to your arms, and possibly legs.

Military and law enforcement personnel are often trained on fighting techniques to handle dog attacks barehanded, such as neck cranks, sharp blows to the throat of the animal, and blinding techniques. I am not sure whether these would be really useful to the average person. In any case, this post is not aimed at military and law enforcement personnel who deal with dog attacks on a regular basis.

Eat beef liver

Beef liver is nature’s super-multivitamin. (Beef heart is just as nutritious.) Dogs, like wolves, have an exquisite sense of smell. If you have seen one of the documentaries about the groundbreaking research by Shaun Ellis (a.k.a., “The Wolfman”), you probably know that wild wolves tend to strongly associate consumption of organ meats with very high status in a pack, to the point that they will instinctively act submissively toward humans that consume organ meats. It is quite possible that dogs do that too. So if you eat beef liver, maybe a dog will “think twice” before attacking you.

Offer the dog a cigarette and a beer

Most dogs can become aggressive from time to time, but not dogs that know how to chill. Therefore, you may consider carrying special dog cigarettes and beer around - only some brands work! Okay, a clarification: the "eat beef liver" advice is not a joke, nor are the others above it.



Notes and acknowledgements

The “charging dog” photo is from Dreamstime.com. The “drunken dog” montage was created with photos from the blog Agrestemundica.

Cesar Millan's site has a number of good suggestions on how to handle dog attacks (). However, I personally think that the way he handles dogs (e.g., often with open hands) is dangerous if copied by an inexperienced person. There is a great deal of "hidden" information that is conveyed to dogs by nuances of Cesar's body language. Those nuances are difficult to copy by an inexperienced person.

An interesting source of information on how to handle dog attacks is the web site Fightingarts.com (, ).

A 2:03:23 marathon world record: Analysis and pacing

Wilson Kipsang's 2:03:23 WR: Thoughts and analysis

Wilson Kipsang yesterday became the 32nd man since WWII to hold the marathon world record.  He broke countryman Patrick Makau's World Record by 15 seconds, setting a new mark of 2:03:23.  A spectacular performance, in which he managed his effort perfectly, showing patience and the right level of aggression at the right time to finish superbly.

Today I share some thoughts, based on the chat on Twitter yesterday, concerning the race, the pacing, and the prospects for that sub-2 hour marathon that people seem very eager to talk about.

I'll tackle the analysis in two parts, one analyzing the race, and later, something on the 2-hour barrier.

The race was managed very well by Kipsang

Below is a graph showing the 5km split times and paces for the race, and includes a projection for 5km based on the final 2.2km, where Kipsang really picked it up.


So, they start fast - the projected time all the way through the first 20km was under the WR.  They reached halfway in 61:32, projecting a 2:03:04, and that was, if anything, perhaps a little too fast.  It set the second half up as a really attritional race - nobody was going to run even or negative splits, and the question was whether any of the big three - Kipchoge, Geoffrey Kipsang or Wilson Kipsang - would hang on well enough to break 62 and the World record.

The pace got slower after halfway.  The section from 20km to 25km was the slowest of the race, which you can see as the peak in the graph above.  Note however that the pace never once dipped below 3:00/km, and that the range was between 2:54 and 2:59.  That's remarkably precision.

That slow 5km interval, and the one immediately after it (25km to 30km) were probably critical to the record because they set the final 10km up.  I have my doubts around how the result would have looked had they continued to push at the 2:03 pace.  It was important to regather, even though it meant slowing down to around 2:06 pace for a short while.

Once regathered, Kipsang was mighty impressive in the final 10km.  He took the initiative, as the "senior" man in the race, and drove the pace even faster.  Only Kipchoge was able to respond and that was fleeting too.  Kipsang ran a few kilometers in 2:49 in that second 30km to 40km, which means there was some in the range of 3:00 too.

So it was a little more varied there, indicating either a change in wind direction (the weather was almost perfect, but not quite), or that Kipsang was digging deep, then finding he needed some active recovery, then digging deep, and recovering, and so on.

Eventually, with 2km to go, he found the big final effort and finished incredibly fast - 2:49/km for the final two, and that was ultimately the difference.  15 seconds, and it came, largely in those final kilometers.  Of course, that's only part of it - the work had been done to get there.

Overall, it was a very well controlled race. Mature, patient, but also aggressive.  His pace-makers did an amazing job up to 30km, and while they may have been just a touch fast from 10km to 20km, it never got out of control.

Taking a broader view, Kipsang paced the marathon almost like a mile race.  If you break the race into quarters, you get the following for his 10km splits:  29:16 - 29:03 - 29:42 - 29:11 (plus that final surge).  The "shape" of that race looks like a typical mile WR - Fast start, slowest in the third quarter, and then the surge.  It was an excellent management of his physiological resources.

In terms of where improvement can come, Kipsang finished very fast, suggesting a reserve, and the ability to go slightly faster.  But it's not huge.  It's not like when you or I finish a 10km and find a surge in the final kilometer that sees us run 20s/km faster than our race average.  In this world, a reserve is being able to go 4-5 seconds per kilometer faster, and so it really is on the limit.

Kipsang is also 31, in his fourth year of marathon running, and set this WR in his 7th race, which is a long time to reach a peak.  Typically, the fastest marathon of an athlete's career happens between 2 and 4, though there are exceptions (Gebrselassie took a while to perfect the race, and then improved steadily quite late).  Kipsang then, may follow a similar approach, and improve again, but the 'safer bet', as it always is, is that he won't.

Behind him, Kipchoge made a big improvement on his debut, which was already impressive at 2:05 from earlier this year.  He's now a 2:04 man, and on the path towards 2:03, so it will be very interesting to follow whether he can continue that, or whether there's a 'glass ceiling'.  The same goes for Geoffrey Kipsang, who "only" ran 2:06:26, but was there for three quarters of the race, and who may yet be able to turn that into an entire race one day.

Then of course there are the Ethiopians, a group of young runners in the 2:04 category, and who may challenge, and there are other Kenyans who've been hovering in the same region.  Marathon running is incredibly deep and strong at the moment, which means we'll get to enjoy similar races and record attempts at least three or four times a year for the foreseeable future.

But for now, it's Wilson Kipsang, with a spectacular performance, who holds the distinction of being the fastest ever.

More later on the 2- hour marathon.

Final thought - the guy who ambushed the breaking of the tape to promote prostitution has been charged with trespassing.  What should happen to him is that he should be sent to Kenya, preferably Eldoret or Iten where all the elite runners train, for three months of community service work.  Let him serve in any was possible (he can carry water and drive behind the runners on long training runs), and learn some respect for the runners of Kenya in the process.  Idiot.

Ross

Sabtu, 28 September 2013

Berlin Marathon 2013: Live splits and analysis

World Record! 2:03:23 Wilson Kipsang
Live Splits and Analysis

Wilson Kipsang has broken the Marathon World Record!  2:03:23 in Berlin.

Here is how he did it, splits and analysis from the race!  I'll post more later!

Overall splits Men

5km:  14:34.  2:55 per km, projecting 2:02:56
10km:  29:16.  14:42 for the last 5km, pace of 2:56/km. Projected time now 2:03:29
15km:  43:45.  14:29 for the last 5km, pace of 2:54/km, the fastest so far. Projecting 2:03:04.
20km:  58:19.  14:34 for the last 5km, pace of 2:55/km.  Projecting 2:03:02
Halfway:  61:32. Easy calculation, it projects 2:03:04, a WR by 34 seconds
25km:  1:13:13. 14:54 for last 5km, pace of 2:59/km, so slowest segment. Projecting 2:03:35
30km:  1:28:01.  14:48 for the last 5km, pace of 2:58/km.  Projection of 2:03:48
35km:  1:42:36.  14:35 for the last 5km, pace of 2:55/km. Projecting 2:03:41
40km:  1:57:12.  14:36 last 5km, projecting 2:03:38. Epic finish coming up!
Finish: 2:03:21.  The World Record is gone!

40km

It's on. Final 2.195km at 2:54/km will get the WR.  By 1 second!  As you can see above, that's what they've done since 30km, and so the record is a real possibility.  Kipsang leads Kipchoge by about 10 seconds, so it is a one-man race for the WR.  Silence now until the end, I'll fill in the blanks later.  This will be a "sprint" through the Brandenburg Gates for the World Record.

Kipsang has slowed slightly in the last 2km, so he needs a pick up.  But it's so close now, if he can just dig in and find 5 minutes of effort, he'll get this.

35km

The pace has now picked up, with the pacemaker having dropped off.  Wilson Kipsang has led the upturn in pace, which has seen the last 5km covered in 14:35 .  That included a 2:52 34th kilometer, very fast.

Wilson Kipsang is the aggressor, leading the race, but with company from Geoffrey Kipsang and Eliud Kipchoge.  Kipsang, as the senior, pedigreed man, obviously has the pressure and obligation to keep the record viable.

With 10km to go, the TV graphic suggests that a 29:30 10km will be needed.  That is definitely feasible.

The pacmaker fought to about 31 km then dropped off, leaving the big three.  It's the Kipsangs, Wilson and Geoffrey, along with Kipchoge.  So as expected, those three fight for the win.  Whether their fight produces a record, that's the intrigue.

30km

They've remained slightly slower than WR pace.  2:58/km gives 14:48 for the last 5km, and a projection of 2:03:48.

So, having been well under WR pace at halfway, it's now going to take a real aggressive final 10km to get the WR.  Whether anyone will take the 'risk' in the company of other men is going to determine how close they get.  That will depend on how they each feel, of course.

At this stage, it's a good time to consider who the viable candidates are.  Wilson Kipsang, Geoffrey Kipsang and Eliud Kipchoge are all there, as is one pacemaker, and Kirwa and Kipchirchir.  An all Kenyan front five, plus the pacemaker Rono.  Five men is good in the sense that 'company' helps in the latter part of the race, but it will be interesting to see how the racing affects the pacing, as it were.

It was this segment where Patrick Makau made the surge that would drop Gebrselassie on route to the current WR in Berlin.  He ran a 5:30 2km segment then, which certainly helped his race, but probably cost him some time.  So the comparison with Makau, which up to now has seen 2013 ahead, will probably look different at 30km, but that's still OK - there are 12km to go from that point, much can happen.

25km

The last 5km were run in 14:54, which is 2:59/km, the slowest segment of the race.  You can tell the pace had slowed because the front group at halfway was thinning out, and it has now expanded again, as runners who had dropped off have come back on.  That's always a sign.  The projection now is 2:03:35, and so it has suddenly come back from being a big WR projection, to a touch and go race.

There is some talk that the runners were benefiting from a tailwind between 10 and 20km, which is now gone.  These are the subtleties that affect WR potential...

At this stage of the race, patience really counts for a lot, so the slowing is not necessarily a bad thing.  The temptation, as the field thins out, is to get aggressive, because you're on the way "home", as it were.  We've seen in London and other big city races how aggression at 25km often blows the race open, but it comes at the cost of the fast time.  So it's important here to be patient, and avoid a 5:35 surge for 2km that can easily derail the WR.  Kipsang of course did that in the Olympics, not off a WR pace, but may have learned from that.  They do have a buffer of around 30 seconds for this second half - a 62:00 still gives a WR.

Halfway

61:32, so a WR projection by 34 seconds.  It promises to be an intriguing second half.  For one thing, the pacemakers will drop at around 30km, and then it will be up to the big three, assuming they're all there, to decide how best to push the pace to keep the WR in view, while still racing and not pulling a colleague to the WR.  That will be perhaps the race's decisive moment.

20km

58:19 at 20km, the projection is for a 2:03:02.  The last 5km was 14:34, so 2:55/km, but there were reports that the 18th kilometer was 2:52, which is very fast and suggests a little bit of oscillation.  Again, the athletes can see their pace and the projected time continuously in Berlin, so when the pace is faster, it's not an accident caused by lack of information, it's a conscious decision to ramp the pace.  They are being incredibly aggressive, and that makes for an interesting second half.  They should hit halfway in about 61:30, and so the second half is guaranteed to be attritional.  The question now is whether it is attritional enough to cost them the WR, or whether they hang on?

15km

The pace has actually increased - 14:29 for the last 5km, and the projected time is now down again, to 2:03:04.  This is quick, and maybe cause for concern.  If you're 15 to 20 seconds up on WR pace through halfway, then that's bordering on reckless.  So it will be interesting to see how the section 25 to 35 km goes.  That's often where the "interest" payments are made.

A TV graphic is showing that they're currently 36 seconds faster than Makau was at the same stage - the coverage is good so far.  Remember that Makau had a race with Gebrselassie that really jacked the pace up after halfway, so that gap may come down later.  The optimal way to run is even pace, so the Makau comparison is less informative, but interesting.  Also, in Berlin, runners have access to the car in front of them, which gives all the information required to manage the pace.  It even gives a projected time, so if they're running under 2:03-pace, then it's because they have chosen to, not because they're making a mistake in the absence of information, which is important to consider.

Florence Kiplagat has gone through 15km in 49:27, which projects a 2:19:06, so she has slowed very slightly, but still on course for a big PB and significant performance under 2:20.

10km

The pace has been maintained, 14:42 for the last 5km.  That's very steady.  The biggest challenge is consistency, so it would be good to see splits by kilometer, rather than 5km, because that would tell you exactly how the pace is fluctuating.  Physiologically, there's a big difference between going 2:52-3:00-2:52-3:00, and running 2:56 every kilometer, even though overall it's the same pace.

So far, that seems to not be the case.  A TV graphic showed a sequence of kilometer splits and the range seems to be narrow - 2:54 to 2:58, so it's a good pacing job so far.  If that continues, then the record is on, and the only determinant is the condition of the atheltes.

No splits from the leading woman, who is Florence Kiplagat.  They're saying her timing chip is not working, so the only splits coming through are for the women in the group behind her.  Will get a split as soon as possible.  She's just gone through 12km in around 39:30, which is 2:18:50 pace, so Kiplagat is going fast too.

5km

14:34, which projects 2:02:56.  The target was apparently 14:40, so they're inside it.  For now, not too damaging (though of course there may have been a 2:40 km in there, I'm not sure), but that is quick.  There's more risk of losing the record by going too fast at this stage.  Not surprisingly, the big three are in the group, along with perhaps 7 or 8 others.  That should thin out at this pace.

Start

As we wait for the first split. a prediction.  I don't think the WR will fall.  Too many things have to be absolutely perfect.  Weather, conditioning of the athlete, the pacing, the intent, and the presence and support of other runners when it counts.  If any of those factors are even 5% below optimal, the price is stiff and the record is gone.

I don't think that the three big names in this race have the necessary ability, so my call is a time just outside 2:04.  Let's call it 2:04:15.  5km split next.