Lately there has been an evolution in the way NHL coaches pull their goalies. Goalies are now being pulled much earlier in games than they used to be. There has been no revolutionary change, however. Goalies are still being pulled at the very end of games, and once pulled they tend to stay pulled until a goal is scored or there is a face-off outside of the offensive zone.
This begs the question of whether any radical change to the strategy of goalie-pulling could still be worth trying. Here are two such ideas:
- Pull earlier than usual to take advantage of exhausted opponents stuck on a very long shift
The risk/reward ratio of pulling a goalie might be much more favourable when an opposing team’s players are exhausted from being stuck on the ice in the middle of an especially long shift.
There are four reasons this may be the case:
First, the ability of players to score into an empty net will decrease when they are exhausted.
Second, the ability of tired players to reach the bench for a line change will decrease when the opposing team pulls its goalie. Typically, players that are exhausted by a long shift are able to escape to the bench either by having their goalie stop play or by getting a lucky bounce off of a missed shot, blocked shot, errant pass, or rebound, which then allows them to clear the puck out of their zone and complete a line change. When playing 5-on-6, however, the likelihood of a goalie being able to force a stoppage of play by making a save decreases, because the goalie’s vision is more likely to be obstructed and the shots that the goalie faces are likely to be harder to save. The likelihood of getting a lucky bounce also decreases, as there is less open ice and there is less likely to be a blocked or missed shot. Plus, even if there is a lucky bounce, it will still be difficult for tired players to clear the zone against six attackers. As such, the tired team may be unable to change lines, and so become even more tired, making the situation even more favourable for the trailing team. And, even if the tired team’s goalie does succeed in stopping play, the trailing team could simply choose to put its goalie back in net for the ensuing face-off.
Third, the willingness of players to even attempt to score an empty net goal will decrease the more tired they become. The reason for this is that if such a scoring attempt fails, it will most likely result either in an icing (which will lead to an even greater tiredness imbalance between the two teams, especially now that there has been a rule change that prevents coaches from using their timeout after an icing) or in an odd-man rush (because tired players will be much less able to get back on defence). As a result, rather than try to score, the tired team may instead focus on trying to get a line change in. During any such line change, the trailing team will usually be able to put its goalie back in if it wants to do so.
Fourth, pulling you goalie can allow you to bring your own best offensive player fresh off the bench, at a time when your own players already on the ice are likely to be somewhat tired themselves.
As a result of these factors, I suspect that it would sometimes be in a team’s interest to pull its goalie earlier than it normally would, in order to capitalize on situations in which the opposing players are exhausted from being stuck on an especially long shift. The question is: how much earlier?
A related question is this: how many goal behind does your team need to be to make such a strategy worth trying at a given point during the second period? Most of the opportunities to score against tired opponents stuck on a long shift occur in the second period, because of the long change.
2. A “5.5-on-3”: introducing a goalie-defensemen hybrid during a desperate two-man advantage
Most 5-on-3 power plays last much less than two minutes long, and on average the odds of a team scoring on a 5-on-3 only become better than the odds of scoring on a typical 2 minute 5-on-4 in cases where the 5-on-3 lasts at least a minute long.
So, not wasting time is crucial. If the penalty killers can kill time by dumping the puck down the ice, or if the power play wastes too much time trying to get the perfect shot (which they often do, as they know that if they do not get a perfect shot, the penalty killers might grab the rebound and dump it down the ice), then that 5-on-3 will be unlikely to score.
In this strategy, then, the goalie on the power play is pulled near the end of the 5-on-3, and replaced with a “safety”: a player who stays near to his or her own empty net in order to protect against long empty net goals, quickly respond to any dumped puck so as to minimize the amount of time the opposing team can waste, and allow his or her own team to avoid wasting time searching for the perfect scoring attempt. This “safety” specialist should excel at being able to serve, in effect, as a player-goalie hybrid.
As an added bonus, by quickly responding to pucks that are dumped or chipped out of the offensive zone, the safety will also make it more difficult for the penalty killers to change lines, which may lead the penalty killers to become fatigued in cases where they have been on the ice for a long time.
The goalie will be put back in the net before the 5-on-3 comes to an end.
In cases where the team is more desperate to score (if they are trailing late in the third period, for example), then this “5.5-on-3” strategy could also be used earlier in the power play.
Another, lesser benefit of this strategy is that, if perfected, it could serve as a sort three-card-monty line change: the “safety” comes fresh off the bench to briefly replace the goalie, then the safety stays on the ice while a more tired power player goes to the bench so that the goalie can reenter the game.
A version of this article was originally published on Rosa & Roubini Associates
When Argentina’s president Mauricio Macri was elected at the end of 2015, investors rejoiced to see the end of the two Kircheners, Nestor (president from 2003-2007) and Cristina (president from 2007-2015). Argentina’s stock market boom, which began in 2013 but faltered just prior to the 2015 election, then continued until this past February. By the start of 2018, Argentina’s stock market had risen tenfold in value in only five years.
Macri had poor timing, however. Just as he was coming into office, commodity prices were falling worldwide. Crude oil prices had been over 110$ a little over a year before he was elected; they fell below 30$ during his first month in office. Soybean prices experienced a similar decline, albeit less sharp of one. This has been painful for Argentina, which is a significant commodity exporter and the world’s largest soy exporter apart from the US or Brazil.
Food exports account for an estimated 62% of Argentina’s total merchandise exports. To put that into perspective, the next highest “food exports as a % of merchandise exports” figure among major economies is Brazil’s, at 32%, followed by Spain at just 16%. Food prices may not have caused the current peso crash — that has been the result of a mix of factors, including the fact that Argentina holds a large amount of USD-denominated debt at a time when the value of the USD is rising. But food prices (and drought) have played a part as well.
The irony, however, is that while Argentina is now struggling as much as almost any country, even having to go back to the IMF which Argentininans have such distaste for, the country is actually uniquely well-suited to facing two of the major, global trends that most pundits and many economists think are likely to occur: rising interest rates and rising tariffs.
The case of tariffs is fairly obvious: Argentina’s exports are equal to only an estimated 11 % of its GDP, lower than any major economy (only the US, Japan, and Brazil even come close). Very few of these exports are to the US. Indeed, Argentina may to an extent even benefit as the US raises tariffs, since doing so leads countries to retaliate by placing tariffs on American food exports that compete directly with Argentinan food exports.
The case of interest rates could similarly benefit Argentina by hurting its competitors’ food exports — in this case, those of Paraguay, Bolivia, and especially Brazil — while leaving its own economy relatively unscathed. Argentina can withstand higher interest rates because it is not a capital-intensive economy; neither in its agricultural sector nor in its non-agricultural sectors. Argentina has low food production costs because its farmland is fertile and has access to ports via navigable rivers. In Brazil, in contrast, farming requires capital-intensive techniques to coax harvests out of acidic, nutrient-poor soils. Those harvests must then be trucked across Brazil’s mountains to reach the sea.
Argentina’s general economy is also not capital-intensive, owing to characteristics particular to its population. To begin with, it s in that goldilocks position of having relatively few elderly citizens or children that require expensively taking care of. It is, moreover, almost wholly urban. 92% of Argentinians live in cities, higher than in any major country apart from Japan or the Netherlands. This can create efficiencies, especially as it is a uniquely concentrated urban population: 38% lives in Buenos Aires. No other major economy surpasses this: the next highest is Tokyo, at 32% of Japan’s total urban population, followed by Paris at 21% of France’s total urban population. With such a large, working-age population, Buenos Aires could be a true economy of scale.
A concievable scenario for Argentina is that rising interest rates will limit soybean and other food production in countries like Brazil, Paraguay, and Bolivia, while at the same time rising tariffs will limit US soybean and other commodity exports. Both will boost Argentina’s export revenues, even as Argentina itself is not hurt very much by rising tariffs or interest rates. If this occurs, Argentina may become a darling of investors yet again.
By far the biggest advantage that a large truck has over a small truck is that a large truck has lower labour costs, per unit of cargo transported. Self-driving technologies that reduce or eliminate labour costs may therefore lead to an increased use of smaller trucks.
This could be particularly likely to occur in areas that have rugged terrain, where labour costs tend to be especially high as a result of slower driving speeds and higher insurance costs.
Small vehicles are also better at handling rugged terrain than large vehicles are. They can make sharper turns, have better control on narrow lanes, can pass through narrower tunnels or overhangs, and can manage steeper inclines.
Indeed, the biggest beneficiaries of automation might be the smallest roads of all: mountain paths that can today only be used by very small vehicles or pack animals. Very small autonomous vehicles could revolutionize transport on such paths not only by eliminating the need to pay drivers’ wages and insurance, but also by gaining more space to carry cargo as a result of no longer needing space for the driver, the steering wheel, and spare tires. These vehicles could be used to facilitate shortcut routes that pass through rugged terrain, or to open up rugged terrain to increased economic activity.
The use of small autonomous vehicles in rugged terrain might also allow for the introduction of another new technology : roll-on, roll-off ropeways. These would be ropeways that small autonomous vehicles would drive on and off of, or clip on and off of, in order to be carried above natural barriers such as steep inclines, rivers, flash-flooded roads, or snowed-in high-altitude mountain paths.
They could be especially efficient at handling inclines, not only by allowing direct as-the-crow-flies routes to replace winding, hairpin roads, but also because ropeways operate as a pulley system wherein the weight of descending vehicles does much of the work — and often does all of the work — of lifting the weight of the ascending vehicles.
Here you can see a very primitive RoRo-Ropeway at work. Here, you can see a somewhat less primitive, though still limited, version built in a Volkswagon factory in Slovakia. If automation leads to a proliferation of small autonomous cars, working ant-like to transport goods in rugged terrain, then perhaps we will see systems like these increase and improve. Economically, it may be as close as we get to flying cars anytime soon.
Roads and railways are great, but they are not portable, scalable, or particularly well suited to ideally handling rugged terrain. You cannot easily disassemble a road or railway in order to move it from location to another. You cannot easily drive trucks or trains up and down steep hills, or across icy or snowy or flooded-out landscapes. And, you cannot widen a road or railway very much without facing sharp increases in expense. The wider you build them, the more likely they are to find a natural or man-made barrier in their way. The widest stretches of highway in the world rarely exceed 150 metres.
Ropeways, in contrast to roads or railways, are portable, are scalable, and are ideally capable of handling rugged terrain. You can fairly easily disassemble and transport them. You can scale them horizontally or vertically. Ropeway corridors could be made extremely wide without being blocked by natural barriers. So long as they are portable and temporary, wide ropeway corridors might also be able to avoid unduly bothering the owners of the private farmland they would inevitably need to cross above at times.
Their potential combination of portability and scale could make ropeway corridors useful for transporting bulk, time-dependent goods: for transporting crops at harvest time, or transporting cargo during the rainy season when roads are flooded out, or in snowy areas during the winter, or to help construct and access mines or dams.
There are a number of factors that could make ropeway corridors become common in the future:
- cheaper intermodal transportation as a result of autonomous cargo transferring. Thus far, the costs associated with transferring cargo from one mode of transport to another have led trucks to account for an estimated 70% of all US cargo transport, despite trucks being generally much less efficient than railways or waterways. If loading, unloading, and handling cargo becomes automated, we might expect that railways, waterways, and perhaps even ropeways will become used more widely as a result
- automating cargo-handling also means that ropeways would be able to add many more entrance and exit points than they have had in the past. Loading cargo on and off ropeways is especially labour-intensive, because the cargo arrives and departs at a slow trickle, each vehicle on the ropeway carrying much less than a truck or train. This has meant that ropeways have tended to move goods only from point A to point B, without many or any intermediate stations. Autonomous cargo-handling could change this, dramatically increasing the usefulness of the ropeway system
- cargo tracking systems — ropeways are slow, which in the past created uncertainties for those waiting for the goods they are bringing. With today’s cheap GPS tracking systems and software that can estimate arrival times (and adjust those estimations when there are unexpected delays of one sort or another), these uncertainties are reduced
- partially-autonomous maintenance: some of the technologies now being deployed or developed for maintaining vast electric grid systems should be applicable for ropeways as well. These include, for example, sensors and computer systems that monitor the entire length of the system in real-time, and drones that can be used as cameras to make inspections
- in some cases passenger cable cars might be able to share the same ropeway as cargo systems – perhaps with passengers being transported during the daytime and cargo transported overnight. Ropeways might also benefit, therefore, from technologies that facilitate intermodal passenger transportation: for example car-sharing, ride-sharing,autonomous valet parking, and other technologies might make it easy for a passenger to drive to a cable-car’s entrance and then get in a different car at the cable-car’s exit.
In democratic countries, political analysts often try to sniff out any regional divisions that exist within a given country by looking at the voting patterns of that country’s electorate. In Italy’s recent election, for instance, it has been thought to be significant that Italians living in northern Italy tended to vote centre/right (including for the Northern League), whereas in southern Italy people tended to vote for the Five Star Movement.
In countries like the People’s Republic of China, however, where no such elections are held, different factors may be looked at instead, in order to gauge the level of regionalism that might exist.
One interesting thing to note here is that almost none of modern China’s top politicians were born in peripherally located areas in the country’s southeast, southwest, northwest, or northeast. The only province in southeast or southwest China to have produced a somewhat notable number of Politburo members is Fujian, a relatively small province where Xi Jinping served 17 years of his career, which is important to China in part because it shares unique social and linguistic connections with the nearby island of Taiwan.
The chart above shows, by birth-province, the number of members in the politburo, politburo standing committee, party secretariat, central military commission, provincial party secretary, or members of previous party secretaries going back to 1990, adjusted to take into account the varying population sizes of each province. Apart from Fujian and Qinghai (which ranks high on this list only because it has such a tiny population, by China’s standards), all of the provinces at the top of this list are in the north or central coastal regions.
China’s most populous province, Guangdong, has had no leaders on this list. As of 2017, Guangdong may have also broken a thirty year tradition by having its provincial governor not be a native of the province. It is now one of the few provinces not to have a native-born governor.
In this picture above, we see the birthplaces of China’s current top leadership, the members of the Standing Committee. Here we also see the one big exception, Qinghai-born Zhao Leji. Zhao’s career is noteworthy. Zhao served as party chief of Qinghai, breaking the unwritten rule that a person should never be party chief in their birth-province. Zhao was later party chief in Shaanxi, where his parents were from; this too broke an unwritten rule, that a person should not be party chief in their “native” province. Now, Zhao has not only reached the Standing Committee, but has taken over Wang Qishan’s anti-corruption job, a critical position. Some have argued that Zhao has been able to rise in this way mainly because Xi Jinping’s family is also from Shaanxi.
Here’s a map of the birth-provinces of the current 25-member Politburo Central Committee (which includes within it the 7 members of the higher-ranked Standing Committee):
And here, finally, is a map of the birth-provinces of the current provincial party chiefs:
Many Conservatives disparage electric cars and bike lanes, while many Liberals fetishize electric cars and bike lanes. The correct approach lies between: some bike lanes and some electric cars are good. Others are not.
For bike lanes, geography can be decisive. Cities like Amsterdam—which is almost entirely flat, and which has no months in which average daily highs exceed 22 degrees celsius or fall below 6 degrees celsius—are ideal for cycling. But most cities are much hillier, hotter, and colder than that. These cities need bike lanes too, but not the same type of bike lane system that Amsterdam has.
For electric cars, size and speed can be decisive. The electric cars currently being marketed to us—the Tesla S, the Nissan Leaf, etc. — are actually far too big and fast to be environmentally or economically efficient. Their batteries expend a lot of pollution during their production, do not provide enough range before needing to be either charged or swapped-out (plus, slow-charging stations, fast-charging stations, and/or battery-swapping stations are all problematic, for various environmental or economic reasons) and are too heavy and bulky to come even close to being ideal.
This is a shame, since electric vehicles in general can be more efficient and eco-friendly than gasoline-fueled vehicles. This is (among other reasons) because they do not contribute to local air pollution, and because they receive their power from power plants, which can be several times more energy-efficient than internal combustion engines and can use energy sources other than fossil fuels.
Electric cars that are much lighter and/or slower than, for example, the Nissan Leaf do not face the same significant battery limitations that electric cars like the Leaf face. If, hypothetically, we all were to decide to buy cars that are closer in their size and speed to golf carts rather than to today’s style of North American automobile, urban areas would very likely experience a substantial economic and environmental gain as a result. The reduced speed limit of the cars would not even cause average driving speeds to drop by much during rush hour, because traffic congestion in urban areas is usually severe enough that vehicles’ average driving speeds already tend to be far below speed limits.
Of course, the goal is not to make people drive tiny cars. Apart from being illiberal, such cars would not be practical or safe on expressways and in suburban areas in which low speed limits would be limiting. The goal, rather, should be to make it safe and comfortable for drivers in urban areas to use small lightweight cars (whether privately owned or, more likely at first, car2go-style rentals), even while they sharing the road with much larger, heavier conventional cars.
Designating certain road lanes (or, better yet, entire streets or downtown cores) as slow-speed limit lanes might accomplish this. Lighter and slow electric cars could safely drive in these lanes alongside conventional vehicles.
Moreover, this could also allow for bike lane systems ideal for cities like Toronto; cities that have a lot of days that are too hot and a lot of days that are too cold/snowy/icy/ to bike comfortably or safely, especially up hills (in summer) or down hills (in winter):
Like electric vehicles, cyclists too would be able to use the slow-speed car lanes relatively safely and comfortably. This could mean three things, all of them good:
- the city would generally be much more bike-friendly than would otherwise be the case
- if you put a two-lane bike lane on one side of the street (see image below), then cyclists would have the option of either using the bike lane or using the slow-speed car lanes — in other words, cyclists would have the option of biking on the sunny side or the shaded side of the street, no matter what time of day it was. This should be very useful on hot days, when cyclists are trying to get to work without breaking a sweat
- instead of having three or four winter months a year in which bike lanes are extremely underutilized, you could instead use the bike lanes during the winter as a parking lane and extra slow speed lane for some of the smaller very small cars (one-seaters or especially narrow 2-4 seaters) that would become common as a result of the slow-speed car lanes. Having a parking lane in the winter would be useful for older people who are at risk of slipping on ice and falling if they have to walk longer distances from their car to their destination.
So, there it is: a plan to promote efficient electric cars, rather than inefficient ones or none at all; and a plan for having bike lanes that could be useful during hot summers as well as during cold winters.
It takes a lot of time to unload a large truck and sort and store its contents. This means that trucks tend to make deliveries during the daytime, when the cost of paying people to unload trucks is relatively low.
If, however, the process of unloading trucks and handling their contents becomes automated, overnight deliveries may become much more common. At night trucks are able to avoid being caught in, and contributing to, traffic jams.
Making more deliveries in the evening or overnight may, in turn, lead to an increased demand for electric trucks. Electric trucks are far quieter than diesel trucks, which is obviously an important trait for nighttime delivery vehicles. They can also be operated relatively cheaply overnight, given the generally much lower price of nighttime power.
If – an enormous if – electric trucks do not need batteries that are heavy, bulky, pollute, and frequently need to be recharged, they can also operate many times more efficiently in general than can diesel trucks.
This is mainly because electric vehicles do not pollute city air, and because electric motors and the power plants that generate their electricty can be several times more energy-efficient (and potentially far more eco-friendly) than internal combustion engines. But it is also because electric vehicles can have regenerative breaking systems that recapture some of the power they expend, and because they have dynamic break systems and motors with very few moving parts, and because they have far stronger torque that helps them climb hills.
Unfortunately, the batteries needed to power trucks are too heavy, bulky, polluting, and range-limited*. This is especially true of batteries for large trucks**, which are the most cost-efficient and eco-friendly types of truck — and which would remain generally the most efficient types of truck even if all trucks were to become self-driving.
[*There may be three main options for dealing with batteries’ limited ranges: slow-charging, fast-charging, or battery-swapping. All three options are problematic. Slow charging is problematic because the nighttime is short, so to spend several hours charging a large truck battery is a waste of precious time. Fast charging is also problematic, because it requires a very large amount of energy at one time, which would then increase peak nighttime energy demand for the grid when lots of trucks are fast-charging their batteries at the same time. If, for example, the wind stops blowing at the same time that many trucks are using wind power to fast-charge their large batteries, power might need to come from fossil fuels, making them much less environmentally friendly. Moreover, if fast-charging stations were used during the daytime too – which presumably they would be, because why spend the money to build fast-charging stations if you are only going to use them at night – it could then lead to increased peak demand in general, which would be both inefficient and environmentally problematic. Battery-swapping stations, then, might be the best option — but building them is easier said than done, given the huge size of truck batteries. Even then, however, they would still not overcome any other issues associated with battery use in trucks.]
[**To quote The Globe and Mail: “Battery powering of heavy duty vehicles may not be expedient. To match the range provided by the diesel fuel tank of a typical long-distance heavy-duty truck, which when full weighs about a tonne, a heavy-duty battery-powered electric-drive truck would have to carry almost 30 tonnes of battery, which is much more than the average payload of heavy-duty trucks.” ]
Barring a breakthrough in battery technology, this only leaves one other option: electric trolleytrucks. These get their power from overhead power wires, somewhat like streetcars do. They then use small batteries in order to travel short distances away from these overhead wires.
Some cities already have large wire-powered networks. Vancouver, for example, which is a city especially suited for electric vehicles given its hilly terrain and cheap, clean, hydropower-generated power, has close to 300 kilometers of wired roads, which it uses for trolleybus transit.
Luckily, trucks making overnight deliveries can avoid the challenges that have thus far prevented trolleytrucks from being commonly used. The main challenge for trolleytrucks has been city traffic. Because they can only travel a few kilometres away from their power wires, they cannot handle the risk of getting caught in stop-and-go traffic.
Overnight, however, the lack of traffic and much longer green light-red light cycles removes this risk. It also means that should a mistake occur that does leave a trolleytruck stranded away from its power wires and out of battery power, it could simply wait for a support vehicle to come and charge its battery, without causing any road traffic blockage as would occur if it ran out of power during the day.
This effectively much extended range away from the wires at night also helps solve another main challenge: lots of people find trolley wires unaesthetic. The ability of trucks to travel further away from the wires at night means you don’t need as many streets wired. You might even be able to get away with only having some highway corridors — where aesthetics is not a problem – wired. The trucks could run on the wired highways during the daytime, then run mostly off-wire overnight to get a few km in the city to make deliveries further from the wired corridor.
A final, hugely significant challenge, which trolleytrucks must face regardless of whether they run during the day or night, is the cost of intermodal cargo transfers. Even if a trolley wire-building spree were to occur, most roads will remain unwired for the foreseeable future. As such, for trolleytrucks to be competitive with diesel trucks, the cost of transferring cargo between trolleytrucks and other vehicles – notably, diesel trucks and trains – must fall. Trolleytrucks being more efficient than diesel trucks will not be sufficient to make them ubiquitous. This can be seen already by looking at the fact that trucks transport much more freight than do railways, despite railways being more efficient than trucks.
If autonomous loading and unloading of trucks, and autonomous sorting and storing of trucks’ cargo, dramatically reduces the cost of intermodal cargo transfers, as seems likely to occur (or at least, plausible), then we might expect the use of cargo railways and of trolleytrucks to increase relative to the use of less efficient diesel trucks.
Indeed, if the automation of intermodal transfers serves to increase
railways’ share of freight transported relative to trucks, one result may be that a larger share of trucking will take place in hilly or urban areas where railways are less competitive. And, since hilly and urban areas are precisely the areas where electric vehicles are most useful — in hilly areas because of their torque, dynamic breaking, and ability to go through tunnels without spewing exhaust that requires ventilation; in urban areas because of their low air and noise pollution – this might further increase the use of trolleytrucks (and trolleybusses!) relative to diesel.