Thursday, July 11, 2019

Clash of Clans War Strategy

Some thoughts about war strategy in Clash of Clans

Clan wars pit an equal number of players from two teams against each other over 24 hours. Each player may attack twice--once on two different players.

For an attack, a player may be awarded 0, 1, 2, or 3 stars.
0. When the player neither beats the town hall nor achieves 50% damage.
1. When the attacker either beats the town hall or achieves 50% damage.
2. When the attacker destroys the town hall and achieves over 50% damage, but not total annihilation.
3. Total destruction of the defender.

A team gets points only for the max stars achieved against each defender. So if one player gets two stars and the next attacker gets 1 star on the same player, the team is awarded 2 stars.

For descriptive purposes, when I talk about player 1 it is the most powerful person on a team. Higher numbers are progressively weaker players.

Strategy During War

Given this structure, it is possible for only the top half of the players to attack and still win.

Think of a 10 player (on each side) game. Maximum points a team can get is 30 (10 x 3).
Player 5 on one team can attack 9 and 10 and likely get full stars 6
Player 4 can attack 7 and 8 and likely get full stars 6
Player 3 can attack 5 and 6 and likely get full stars 6
Player 2 can attack 3 and 4 likely get partial stars 5
Player 1 can attack 1 and 2 and likely get partial stars 3
Total 26 stars.

I have seen this done and it works. It has a few flaws though. It requires perfect attacks by each player. There is no room for network lag or an accidental tap. There is no room for error. Also, if the lower players are not just shell players, it's boring for them.

Better then is a phased approach.

1. Bottom five get in their attacks. Best if you can coordinate 10 - 6 in order, but that's difficult.
They should use a similar pattern and not worry about beating better players. So 10 can attack 9 and 10, 9 can attack either of them if 10 doesn't 3-star them and so forth.
2. By the time the bottom half attacks, they should have most of the bottom stars done. Possibly only 6 and 7 will have a star remaining.
3. Then the powerful players step in. In a similar fashion, 5 can clean up the lower players with remaining stars, then 4, etc.
4. With an hour left in the war, the top two or three players can take the actions needed to get the most stars.

I believe this phased approach is the best one for maximizing stars if you can coordinate, which, let's face it, is difficult. This approach also allows for multiple attacks on enemy bases if they are difficult or tricky.

So here are the rules you need.

1. Higher numbers (bottom half) use your attacks in the first 12 hours.
2. Lower numbers try to work together to sequence and maximize stars
3. Everyone use both attacks. If your first attack fails, hit a lower number player. Everyone should be trying to 3-star at least one opponent.
4. If you are a strong player and you need to attack early because real life, do NOT attack somebody that is more than 2x your number. So 3 can attack 5 or 6, but not 7 - 10.

Filling Defense Castles

Let the highest players fill castles. Put powerful defenders in all castles. If the opponent is forced to waste strong attacks on weak players, that's good for us.

I won't comment on the best layout or troop donations, there are tons of resource out there. Here's one on the best troop donations

Wednesday, March 13, 2019

Voting Systems

Ranked Choice (RC) vs Instant Runoff (IRV) vs Approval

RC and IRV are technically different. RC is a way of expressing preference, IRV is a method of tabulating votes from RC ballots.

This is an important distinction.

IRV, which seems pretty good can solve one problem (e.g. Gore v Bush) but introduce others. Technically IRV is nonmonotonic. That is, a shift of public opinion toward a candidate can cause that candidate to lose, and a shift of public opinion away from a candidate can cause that candidate to win.

A different and probably superior method is one called Condorcet. It eliminates most of the IRV issues but introduces a few other, but less likely, issues. IRV is to Condorcet as single elimination is to round robin.

Superior to both of them is Approval voting, in which you select all candidates that would be acceptable to you. It’s simple to tabulate and does not have the issues of either IRV or Condorcet.

There’s a website that has wonderful graphics of how these different systems can work or not depending on the positions of the candidates.

The author of the site is Ka-Ping Yee.

Our election method makes it virtually impossible for third parties to gain a significant foothold. The electoral system is an added complication for presidential races, but the same holds true for state and local races as well as primaries with crowded fields.

Monday, January 07, 2019

Portfolio Balance, Investment Strategy, and Sources of Retirement Funds

I'm hopefully about four years away from retirement. I have a good salary and, looking out to 2023, I think I'm in good shape overall. There will be three main sources of funding for the retirement years.

1. Wife's and my 401ks and IRAs
2. A lump sum retirement contribution from my employer
3. Social Security

1 and 2 are about equal right now, 3 is a fixed future monthly payment.

The stock market has been hot since the crash of 2008. The Dow is now around 25,000. How much higher will it go? Is it going to crash? Would it be wise to transfer some funds from stocks into cash or bonds?

My first thought was Yes, of course! But, I did a little digging on this and one piece of advice was that people approaching retirement should have a couple years worth of cash in their accounts so they don't have to sell stocks in a down market. That makes a lot of sense, I have seen literature that shows how a stock market decline immediately following retirement can demolish retirement plans.

But then a realization hit me. My lump sum and social security future payments are, in essence, cash funds. They have little risk and little potential for appreciation. So I have the cash reserves available at retirement. In fact, with this realization, I am quite overbalanced in cash. My 401k portfolio has always been on the "risky" side, i.e. geared for appreciation. Financial advisors through the years have advised me to ratchet down the risk and invest in bond funds, etc. This didn't make sense to me--mainly because with interest rates at 0%, bond funds were likely to decline in value. Yes, you get interest payments, but that gets offset by value depreciation as interest rates increase.

So this realization that I already have over 50% of my retirement funds in cash was freeing. I consciously realized for the first time that my "risky" investment strategy was actually rational and, in fact, quite conservative.

I understand that defined benefit retirement programs (like my lump sum) are increasingly rare in the corporate world today, but people do get inheritances, life insurance, and other possible (usually sad) surprises as well. So this is a way to think about the potential for those as well.

And by the way, when you are highly confident that the market is heading for a fall or is very low (i.e. it is at a bookend, you should consider making some moves in the buy low sell high framework. Don't bet everything on a market turn, but don't ignore it either.

Tuesday, July 31, 2018

Donald Trump is not my vision for America

What trump is doing.

Two things:

He is fanning the flames of division, essentially betting that if his base, approximately half of the electorate has enough fear and motivation, they will vote him in. He is not trying to cross any divide or reach across the aisle. He is attempting to rule by getting a slim plurality to hate and fear the other half. He is succeeding. He is treating the US as a “majority rules” system, which we are to some extent, but that is not the fundamental principle of America. We are grounded by a constitution founded on the rule of law. Our greatest hope is that some portion of those that support trump today realize what he is doing and how it is undermining our nation.

I read an article recently that questioned how people could be loyal to trump, when he shows no loyalty towards them. Interesting perspective. It made me think though back to some of the management training or reading I have done over the years. Specifically the PIC-NIC framework for motivating people. PIC is Positive Immediate and Certain; NIC is Negative Immediate and Certain. Carrot and stick. Once you go outside the immediate and certain side, the motivators are much less motivating. Trump’s style is to emphasize the NIC. He rules largely by fear. It is a powerful style. But in the case of his former attorney Cohen, he no longer has the ability to provide short-term reward or punishment (within the law). Maybe he could give him a pardon at some time in the future, but that then gets outside of the IC framework. It is the future and uncertain. I hope the FBI is protecting Cohen well, otherwise, he is likely to receive a visitor with some NIC motivation.

Thursday, May 17, 2018

Sea Level Rise

Sea-level has undeniably been rising for the last 10,000 years, since the end of the last ice age. In the 130 years from 1870 to 2000, global sea level rose by about 200 mm. That’s 8 inches. This works out to about 1.5 mm/yr. Between 1950 (when the anthropogenic signature begins) and 2000, sea level rose about 75 mm (3 inches).

That’s a pretty decent historical record. 

A simple linear extrapolation of this trend gets you to 125 mm by 2100. That is an additional 5 inches or so. I’m not saying that is a good estimate of what will happen, but it’s probably a good starting point. In my view, it is probably towards the low end of what we would expect to see in the next 80 years. In a bay with 6-foot swings in tide twice daily, 5 inches is not going to change anything.

According to satellite data since 1993, the rate is about double the historical rate. If that trend were to continue, sea level will be another 10 inches above today’s level in 2100. This is about the same rate as sea level increased from 1930 to 1950 (before the anthropogenic signature), so this is not an extreme event.

Anything beyond that 5 inches has a high burden of proof on it. The IPCC does have scenarios that have numbers like what you have suggested 2 meters, but there are a lot of worst-case assumptions built into that.

So should we assume 10 inches and stop worrying? No! First, low-lying areas need to be smarter about development. Any area whose infrastructure is less than the high tide level is relying on pumps for drainage and sewage. The good news is that we have proven it can be done in Europe and New Orleans. The bad news is that it is energy-intensive and costly to maintain. sea-level measurement includes adjustment calculations, e.g. what would the sea-level change have been if the ocean basins had not increased in volume. (WTF?) If we assume that the trend before 1950 was natural (we really did not emit much CO2 into the atmosphere before then) and that the following increase in the trend since 1950 was 100% due to humans, we get a human influence of only about 0.3 inches per decade, or 1 inch every 30 years.

Wednesday, December 27, 2017

Previous Hypothesis Disproven: Trump is NOT ethical or moral

I recently wrote a post as a thought experiment—Hypothesis Trump is moral and ethical. In a nutshell, that hypothesis is disproven.

Some people read my headline and assumed that I am pro-Trump. I am not. I have never believed that Trump is a good person. His actions in the White House have demonstrated that enriching himself and his family is a high priority of his administration. At the same time, I know I live in an echo chamber that accentuates the negatives about Trump, and seldom presents different possibilities.

In my view, the possibility existed that most of Trump's accomplishments had been relatively harmless signals to his constituency. Many people believe that Trump has caused a lot of damage tot he country (including reputational), despite lack of significant progress, and I get that. I think that until now, most of the damage has been easily reversible. I selected this issue as the litmus test, because there was pretty widespread belief among economists and laypeople, that the bill would be a net negative, it was clear that Trump would benefit from the passage, and it was the first significant action he had a chance to take. Remember, repeal of ACA never got to him.

In my mind, there was a possibility (maybe a hope), albeit slim, that Trump had good intentions with respect to the presidency and the nation. Scott Adams argued that Trump first needed to take leadership of the Republicans by pacing and leading. In other words, prove that he is one of them, then use that identity to steer the direction (see point 5 in this link).

Some of my friends thought I had gone into "engineer mode" when I put the previous post out there. I had. It is my best tool to try to overcome confirmation bias.

I ended that previous post with the following statement, “If he signs the bill, we must reject the hypothesis that he is a moral and ethically aware adult. If he vetoes it, there is still a possibility that he is morally and ethically aware.”

Any belief that Trump has a shred of decency is now disproven by his signing of the bill. The bill was passed purely along party lines, against public opinion polls, and despite the fact that every non-partisan analysis showed that it would increase the deficit, reduce aid to the needy, and transfer tremendous wealth to the top few percent of the nation. Furthermore, I can't think of any larger conflict of interest in history for a person elected or appointed to oversee the public trust.

Sure, many people will see a lower tax bill in 2018, but over the next 10 years, those tax savings decline. Personal tax savings for middle-class workers completely goes away in 2027, when their tax cut expires. On the other hand, Trump and his family will reap tremendous personal benefit from the new tax code. Keep in mind that a 10% savings for someone paying a 10 million dollars in tax is 1 million; A 10% savings for someone paying 10,000 is only 1000.

By the way, if anyone wants to argue that because the bill does not eliminate the estate tax, Trump dd not benefit much, I have a few points to make. First, he will get plenty of benefits from the change in the pass-through tax rate as well as other benefits granted for real estate businesses. Second, Trump would not personally benefit from the estate tax, his family would. Finally, this game is not over. He has three years to take other bites at this apple. Repeal of the estate tax will come back. For anyone saying it is double taxation, I have one thing to say (which will not change your mind). The deceased is not double taxed. He is taxed once when the money is earned (maybe), then when he dies his family is taxed on their receipt of the money.

The bill is also balanced in a way that punishes blue states while enriching red ones. Trump's signature is a completely unethical money grab by him and his fellow 1%ers. Trump is not draining the swamp, he is the swamp.

Monday, December 04, 2017

Hypothesis - DT is ethical and moral

Background: The tax bill recently passed by Congress will have two results: benefits will go down for the majority of Americans and taxes will go up for the majority of Americans. Donald Trump and his family are in a position in which they will overwhelmingly benefit. They do not need most of the benefits, and DT's inheritance taxes alone will decrease by $4 billion. His family will also benefit enormously from the change in the pass-through rate. I am assuming that he pays taxes in the US at all, which based on the one tax return he made public is the case. It is fair to assume that the Trump family will benefit by more than $5 billion personally overt he next 10 years or so.

Hypothesis: Donald Trump, as president of the United States is in a position to approve or veto a bill that will benefit him more than 99.99% of his constituents. Human beings all have a self-serving bias, and ethically and morally aware adults realize this. As a result, if Donald Trump is ethically and morally aware, he will veto the bill and force Congress to pass it with a 2/3s majority, thereby ensuring that the right thing is being done for America.

If he signs the bill, we must reject the hypothesis that he is a moral and ethically aware adult. If he vetoes it, there is still a possibility that he is morally and ethically aware.