Mahjong Average Placement | Calculation, First-Place Rate, and Last-Place Rate

Is average placement2.4 strong? Is 2.3 a pro level?

"average placement", which is often used for mahjong results, is the average of the final rankings over a certain period of time. The mathematical center is 2.50 for four-player mahjong and 2.00 for three-player mahjong, and the smaller the number, the more games you finished in the top.

If you maintain average placement2.40 for hundreds of games with four-player mahjong, it can generally be evaluated as a strong performance. 2.30 units is very good. However, since it is average placement2.30, it cannot be called "professional level" right away. This is because the strength of the opponents differs between regular tables, advanced tables, and professional leagues, and the difficulty level of achieving the same number also changes.

Also, even a difference of 0.05 for average placement will result in a total difference of 25 points in 500 games. In this article, we will explain in order the guidelines for four-player mahjong and sanma, the standards from beginners to top class, examples of M Leaguers, the number of games required, and ideas for improvement.

1. What is average placement of mahjong?

1-1. average placement is "average final ranking"

average placement is obtained by adding the 1st, 2nd, 3rd, and 4th place numbers for each game and dividing them by the number of games. The smaller the number, the better the result, and it shows the overall result including 2nd, 3rd, and 4th place, which cannot be seen only by first-place rate.

1-2. How to calculate average placement

There is no 4th place in sanma. For example, if 1st place is 38 times, 2nd place is 33 times, and 3rd place is 29 times, average placement is（38＋66＋87）÷100＝1.91It is.

1-3. What can we learn from average placement?

With first-place rate alone, it is difficult to distinguish between "those who finish many times in the top but often in the last" and "stable types who often finish second", and with last-place rate alone, it is difficult to tell how to rank high. average placement combines all the finishes into one, so you can intuitively check whether you finished in the top overall position.

2. average placement of four-player mahjong is based on "2.50"

2-1. Why is the standard for four-player mahjong 2.50?

If we take 25% equally from 1st to 4th place, the average is（1＋2＋3＋4）÷4＝2.50It is. Assuming that everyone has played enough games with exactly the same ability, the score should theoretically converge to around 2.50. Therefore, under the same environment, a score below 2.50 is better than average, and a score above 2.50 is considered to be struggling more than average.

2-2. four-player mahjong Basic quick reference table of average placement

This table is a rough guide if you play against opponents of the same level in a sufficient number of games. Evaluation changes depending on rank, table level, and rules.

3. average placement of three-player mahjong is based on "2.00"

3-1. Why is the average 2.00 standard for sanma?

sanma only has 1st, 2nd, and 3rd place, so if it is even（1＋2＋3）÷3＝2.00It is. sanma's 2.00 is an average position, while four-player mahjong's 2.00 is a very good score. Even though the numbers are the same, the meanings are completely different.

3-2. Basic quick reference table of average placement of sanma

4. average placement of four-player mahjong and sanma cannot be directly compared.

4-1. four-player mahjong2.00 and sanma2.00 have completely different meanings

four-player mahjong2.00 is an amazing number that is 0.50 better than the standard, but sanma2.00 is the center of mathematics. Be sure to separate and tally scores for different people.

4-2. Compare the standard values of four-player mahjong and sanma

4-3. Think in terms of “how far is it from the standard?”

four-player mahjong2.40 and sanma1.90 are both 0.10 better than the center. However, this is just a convenient way to grasp the positional relationship. Since the rules, game speed, and ranking distribution unique to sanma such as tsumo loss and North extraction are different, it cannot be used for complete skill conversion.

5. [four-player mahjong] How strong should average placement be? Estimated ability level

5-1. 2.60 or higher | I want to aim for 2.50 first

If average placement is over 2.60, it may not mean that you are not in the top position, but rather that you are in 3rd or 4th place. First, let's open the rank distribution and determine whether only last-place rate is high or whether top-two rate itself is low.

Things that are likely to lead to improvement include pushing too hard on your opponent's riichi, forcing yourself to play with low-scoring hands, and proceeding without checking the finishing conditions in your auras. Rather than trying to fix everything at once, it will be easier to find the cause by focusing on the situations you look back at, such as ``the station that received the preemptive riichi'' or ``the station that finished 3rd or lower in the southern field.''

5-2. 2.50 to 2.59｜Zone around the average

2.50 is the mathematical center of four-player mahjong. Think of the 2.50 range as a range where it is easy to move up and down with slight deviations between first and last, second and third place, rather than not being able to win. If the number of games is small, it may deteriorate to nearly 2.60 after just a few games.

For beginners, getting below 2.50 using the same rules and table level is an obvious first goal. Rather than trying to get to the top every time, average placement will improve more stably by getting off safely from uncontestable moves and getting one higher rank in auras.

5-3. 2.45-2.49｜Above average

average placement2.45 to 2.49 is the zone where the value has started to fall continuously below the standard of 2.50. The difference in numbers may seem small, but if it is 2.45 in 500 games, the total number of positions will be 25 fewer than if it were 2.50. If it can be maintained over a long period of time, it is a good result that is difficult to explain by chance alone.

From here on, it will be important to not only reduce simple mistakes, but also make detailed decisions to move from 2nd to 1st, or from 4th to 3rd. By lining up first-place rate and last-place rate and seeing which side has room for improvement, it will be easier to decide on the next issue.

5-4. 2.40-2.44｜Line that is easy to say “strong”

The low 2.40 range is a level that can be easily evaluated as "strong" in four-player mahjong. If it is 0.06 to 0.10 better than the standard and has been maintained over several hundred games, it is considered that you have stable results not only in tile efficiency but also in pushing and pulling and judging the order of finish.

However, the 2.40 recorded at a beginner's table and the 2.40 recorded at a table attended only by advanced players are not the same difficulty level. When telling people your numbers, it is difficult for people to misunderstand your ability if you include the aggregation conditions, such as ``four-player mahjong, Hanshu, Advanced Table, 2.42 for 500 games''.

5-5. 2.35 to 2.39 | Quite strong

2.35 to 2.39 is a guideline for the advanced class. Even just a 0.05 drop from the average of 2.40 can result in a difference of 50 points in total rankings over 1,000 games. If you stay within this range for a long period of time rather than a short-term winning streak, you're doing pretty well.

At this level, it becomes difficult to maintain either the offensive power that increases the top players or the defensive power that suppresses the last players. It is important to look at the ranking distribution and tiles, and look back at the 2nd place that missed out on the top and the 4th place that could have been avoided.

5-6. 2.30～2.34｜Very strong

If you can maintain the low 2.30 range over hundreds to thousands of matches, this is a number that can be called top class in a general competitive environment. This is a level that is difficult to reach unless you consistently stack up 1st and 2nd place and also suppress the fatal last.

On the other hand, you cannot conclude from average placement alone that it is "stronger than the pros." The strength of the opponents is different in general tables, top tables in ranked matches, and professional leagues. The value of a 2.30 result depends on the environment, against whom, and how many games you hit it.

5-7. 2.20 units | Top professional level numbers are coming into view

average placement2.20 units are extremely excellent as a long-term performance for general players. Even in league matches between professionals, the numbers are at a level that is recorded by top players who are doing well on a season-by-season basis. This is an eye-catching number even for 100 games, and the value increases even more if it continues for several hundred games or more.

However, when the number of games is small, the number of consecutive wins and last matches is strongly reflected. Don't just look at the glamor of the numbers, check first-place rate, last-place rate, opponents, and aggregation period, and think about short-term success and long-term ability.

6. [sanma] How old should average placement be? Estimated ability level

6-1. 2.10 or higher | There is a lot of room for improvement

If sanma is 2.10 or higher, there is a possibility that the games that will result in 3rd place overlap. sanma moves quickly and tends to result in high-scoring exchanges, so a single shot or a point conceded on dealer often directly affects the final ranking.

Start by checking the third-place rate. Review hands where you lost many points as dealer, pushed too far against another player's riichi, or missed a realistic placement improvement in the final hand. The correct decision depends on the rules, so record whether North extraction and tsumo loss were used.

6-2. 2.00 to 2.09 | Around average

2.00 is the mathematical center of sanma. In this range, average placement will make a big difference if you finish first or third a few times, so first check the number of matches. 1.98 for 20 games and 1.98 for 500 games have very different meanings.

The first goal is to consistently score below 1.99 with the same rules and table level. In addition to increasing your attack power, you will also be able to improve your average placement by not forcing yourself too much in uncontestable positions and reducing the number of big third-place finishes.

6-3. 1.95-1.99｜Above average

The number of games that are below the standard of 2.00 and end in the top ranks is increasing. It's not uncommon for a short period of time, but if you can maintain it for a few hundred matches, it can be evaluated as an above-average performance.

At this stage, let's check whether first-place rate is increasing or whether the third place rate is decreasing. Even if the result is the same, if you understand what needs to be improved, you can move on to the next challenge while preserving the strengths of your own hitting style.

6-4. 1.90～1.94｜Strong

The low 1.90 range is a level that can be easily called "strong" for sanma. If it is 0.06 to 0.10 better than the center and stays within this range over the long term, it is considered that the results reflect both the ability to take the top spot and the ability to avoid third place.

However, sanma is a game with large rules differences. Please do not mix results with different tsumo loss, North extraction, number of red tiles, ranking points, etc. and compare them under the same conditions.

6-5. 1.85 to 1.89 | Quite strong

1.85 to 1.89 is a guideline for advanced class. If you have a record of more than a few hundred matches, it is difficult to achieve just a winning streak, and you need to be able to reduce the number of third-place finishes even in the face of bad tile placement or poor game play.

In online mahjong, the higher the table level, the more difficult it is to maintain the same number. If you include not only average placement but also your current rank and table, you can accurately convey the value of your results.

6-6. 1.80-1.84｜Very strong

If you can maintain it in the low 1.80 range over a long period of time, that's very good. Not only is first-place rate high, but there are many cases where the third place rate is also kept low, so let's open up the ranking distribution and check the details of the results.

Even with 50 or 100 games, this is a respectable number, but sanma's average tends to change after one big winning streak. It is important to see whether the same trend remains even when the number of matches is increased to 300 or 500 matches.

6-7. 1.70 units | Top class

1.70 units is top class in a general competitive environment. Even in the short to medium term, this is a very noticeable result, and if it can be maintained over the long term, it is considered to be a reflection of high ability.

On the other hand, the numbers tend to be extreme for tables with a small number of games or large differences in skill, so be sure to check the results for ``how many games'', ``which table'', and ``what rules''.

6-8. There is no “professional common standard” for sanma

sanma does not have a typical environment like four-player mahjong's M League where many top professionals compete under common rules and publish individual results in the same format. Rules and participant demographics vary depending on the competition organization, tournament, and online service.

Therefore, it is not appropriate to set a uniform standard such as "average 1.80 or less for professionals." For online mahjong, add rank and table level to average placement, and for tournament results, evaluate participants, number of matches, and recruitment rules.

7. How would you summarize average placement as "beginner, intermediate, or advanced"?

7-1. average placement guideline by ability of four-player mahjong

7-2. average placement guideline by ability of sanma

The actual strength varies greatly depending on the number of games, opponents, rules, and table level. The table is not an absolute dan recognition, but a guideline.

8. What average placement do professional mahjong players achieve?

8-1. There is no absolute standard that says “average placement○○” if you are a professional.

There are differences in ability among professional mahjong players, and in league matches, the opponents are also professionals. 2.35 at a regular table and 2.35 at a professional league are not the same value. You cannot judge whether you are a professional or not just by looking at average placement.

8-2. When you watch the M League, you can see how great the top professionals are

The M League is a four-player mahjong league where top professionals gather.2025-26 regular season official resultsHere, the top players recorded average placement2.1 to 2.2. On the other hand, there are also players with 2.5 or 2.6 cars or more. Since it is a relative evaluation between top professionals, it is not "a professional is always below 2.3".

8-3. average placement example for 2025-26 M League top players

The M League official describes average placement as "plain clothes". The numbers and number of games are displayed on the official 2025-26 regular season page.

8-4. Outstanding performance in the M League with an average of 2.2 units

In an environment like the M League where everyone is at a high level, it is difficult to fall far below average placement2.50. Rather than just comparing the numbers with the 2.20 range at general tables, you need to evaluate it by including the fact that it is 2.20 against the top pros.

This also applies when looking at your grades. Instead of just treating average placement like a profile, you can only understand the value of the results by checking who the numbers are for.

8-5. Separate season results and long-term results

Individual appearances in the 2025-26 regular season range from 20 to 39 games. Hitting the 2.10 range over 30 to 40 games is great, but it doesn't guarantee you'll be able to maintain that number for hundreds of games.

On the other hand, even if average placement is bad in one season, it does not mean that its long-term ability will be denied. Even for professionals, their results fluctuate greatly from season to season, so read the rankings for a single year and the results for multiple years separately.

9. Where does the “professional level” average placement come from?

9-1. Estimated four-player mahjong based on numbers only

9-2. “average placement2.3=Professional level” does not necessarily mean

2.30 at a general table, 2.30 at an online mahjong advanced table, and 2.30 at a professional league are the same numbers, but the difficulty to reach them is different. This is because the higher the opponent's ability, the more difficult it will be to lower average placement significantly from 2.50.

On the other hand, even if you hit only against strong opponents and hit around 2.50, that alone doesn't mean you're weak. average placement is a relative result within the group to which you belong, so avoid cutting off the numbers and using them for professional certification or ability certification.

9-3. The true professional standard is “numbers + competitive environment”

If you want to evaluate the strength of a professional, you need to check not only average placement, but also the number of games, opponents, competition rules, and long-term reproducibility. In rules with large ranking points, the value of the top position increases, and in rules that emphasize avoidance of the last position, the value of defensive decisions increases.

In addition, promotion in the professional league, advancing to the title game, and point balance over multiple years are also important. For the first time, you can see your overall strength as a professional, including whether you can continue to produce results in different environments.

9-4. Even professional mahjong players have negative seasons.

mahjong has random fluctuations, so even top players experience positive and negative seasons. In league matches of dozens of games, in addition to skill, the effect of tile placement and development is also greatly reflected in the results.

Just because average placement was bad in one season, it doesn't erase the previous achievements. The strength of a professional is determined by his ability to maintain good results over many years against strong opponents, and by making adjustments even during bad times and producing results again.

10. If you look at M Leaguers as a whole, you can understand the meaning of “2.50” even more

10-1. Even among top pros, the overall score is around 2.50.

In four-player mahjong, the total ranking for each match is always "1+2+3+4=10". Dividing it by 4 people gives us 2.50, so if we look at all the games in the same league and weight them appropriately by the number of games, the overall average is 2.50.

In other words, for someone to record 2.20, someone else's average placement is over 2.50. Even if all the participants are top pros, it is mathematically impossible for everyone to get below 2.30.

10-2. Even M Leaguers have seasons where the average is 2.6.

In the 2025-26 regular season, there are multiple players with average placement of 2.6 or higher. However, it is not appropriate to interpret this in the same sense as 2.6 for general consoles and judge that "mahjong is weak."

In the M League, the opponents are carefully selected professionals, and each player has appeared in dozens of games. Short-term results against strong opponents can show harsh numbers, so it is also necessary to check long-term results, including past seasons and other leagues.

10-3. average placement is a relative evaluation including the environment

average placement is an index that easily shows how high you rank among the same group. On the other hand, the strength of the group itself and the difficulty of the rules are not included in the numbers.

When looking at your own performance, just because your average placement worsens immediately after moving up in rank does not necessarily mean that you are weaker than before. If the competitive environment has changed, do not simply compare the numbers before and after, but instead evaluate the same environment separately.

11. Looking at professional results, average placement alone is not enough

11-1. Many metrics are published in the M League

On the M League official results page, in addition to individual scores and uniform performance, first-place rate, top-two rate, last-place avoidance rate, average batting average, shooting rate, firing rate, open hand rate, riichi rate, etc. are published. The reason there are so many indicators is that a single number alone cannot explain a player's performance.

For example, if you want to know why average placement has gone up, check which one went up: first-place rate or last-place avoidance rate. If you want to look at the batting side, read the attack rate and average batting average, and if you want to look at defensive trends, read the firing rate as well.

11-2. Same average placement but different contents

Even with the same 2.40, there are three types: an aggressive type that takes many tops and lasts, a stable type that avoids the last and stacks up second place, and a type that has a relatively even finish order. average placement compresses these distributions into one, so it does not directly represent the characteristics of the hitting style.

Also, even if you can guess how to play based on your score, you won't know if your individual choices were correct unless you look at the tiles. It is important not to conclude that everything is pushed too hard because the firing rate is high, but to go back and consider the round, the value of the tiles in hand, and the point stick situation.

11-3. Individual scores are different even if average placement are close

In addition to finishing points, the M League individual score also includes raw points, which is based on the points you have at the end of the game. Therefore, even if you are in the same place, you will earn different points for a big top and a close top.

Even if two athletes are close in average placement, if one of them has many big tops and the other has a lot of close second places, there will be a big difference in their individual scores. average placement is another indicator that reflects ``what ranking you placed in'' and points reflect ``what difference in points you finished with''.

12. Even though it is the same average placement2.40, the contents are completely different

12-1. Comparison of top type and last avoidance type

12-2. average placement is an index that combines rank distribution into one

A has many tops and lasts, while B has few tops but hardly any lasts. Even so, if you multiply each rank by the rank number and add them up, both will be average placement2.40.

If the rule has a very high value at the top in terms of ranking points, A's performance may be advantageous, and if the rule has a heavy penalty for the last place, B's stability may be evaluated. Even if average placement is the same, the desired distribution will vary depending on the rules and objectives.

12-3. Also see first-place rate and last-place rate as a set

When average placement improves from 2.50 to 2.40, check whether the number of tops has increased or the number of lasts has decreased. If it's the former, his decision-making ability has improved, and if it's the latter, his defense and finishing order management may have stabilized.

Furthermore, if you look at the 2nd and 3rd place rates, you can see which way the narrow difference in rankings could have been moved. By using average placement as an entry point and proceeding to the ranking distribution, you can narrow down specifically the games you should review next.

13. Is the difference between average placement0.05 big?

13-1. Is there only a 0.05 difference between 2.50 and 2.45?

It may seem that there is almost no difference between 2.50 and 2.45, since the only difference is in the second decimal place. However, by multiplying the difference in average placement by the number of games played, you can see how much the total ranking has changed over the entire period.

For example, for 100 games, a 0.05 difference is a total difference of 5 positions, and for 1,000 games, a total difference of 50 positions. If 4th place changes to 3rd place, the total ranking improves by 1, so over 1,000 games, such changes will be accumulated 50 times.

13-2. What about 100, 500, and 1,000 games?

13-3. For average placement, "decimal places" are important

With average placement, the longer the difference after the decimal point, the more significant it becomes. However, a difference of 50 in total rankings does not directly represent a pure difference in ability. It is also necessary to check whether the opponents or rules have changed during the course of the game, or whether you have won significantly over a specific short period of time.

Don't overestimate the difference in short-term comparisons such as month-on-month comparisons, and carefully track changes of 0.01 or 0.05 in overall results. By using these two perspectives, you will be less likely to be swayed by numbers.

14. How many games are needed before average placement is reliable?

mahjong has large short-term fluctuations, 10 to 20 games are only for reference, and even 50 games still fluctuates significantly. You can start to see trends after 100 games, and from 300 to 500 games it becomes much easier to refer to your long-term results. Even after 1,000 games, your ability is not completely determined.

There is no strict common line on the number of games required. Please use this table as a guide, and be aware of confidence intervals and reproducibility under the same conditions.

15. Which is better: “average placement2.2 for 100 games” or “2.35 for 2,000 games”?

average placement2.20 for 100 games is an overwhelming short-term result. By anyone's standards, it was a good period with many tops and few lasts. However, in 100 games, the effects of tile distribution, opponent bias, and winning streaks will still be significant.

On the other hand, maintaining a score of 2.35 over 2,000 matches may not be as impressive as 2.20 in the moment, but it has a repeatability that is significantly above average over a long period of time. As a material for estimating your ability, this will likely carry more weight in many situations.

Of course, 2.20 for 100 games and 2.35 for 2,000 games are both amazing results in different ways. Don't use the same yardstick for short-term achievements and long-term stability, and be sure to write the number of games played next to it, such as ``2.20 (100 games)'' or ``2.35 (2,000 games)'' when evaluating.

16. How do you view average placement on online mahjong such as Jakukon and Tenho?

In online mahjong, the value of the same average placement changes depending on the table you are playing on. This is because it is generally more difficult to maintain the latter of 2.35 at a table with many beginners and 2.35 at a table with only high-level players. When viewing the results screen, consider the dan and table levels as part of average placement.

Immediately after moving up to a higher level, average placement may become worse than before. This is not necessarily a result of a decline in ability, but rather a result of the opponent becoming stronger. If you separate the records from different tables into one before and after promotion, it will be easier to judge whether you are adapting to the new environment.

If possible, please separately tally not only four-player mahjong and sanma, but also Dongfeng and Hanzhuang matches. Different services like Jakukon and Tenho have different matching, tier systems, and recruitment rules. The basic idea is not to treat the same 2.35 as having the same value, but to compare it with your past self within each service.

17. If you are going to watch average placement, you should also check the results

17-1. average placement＋first-place rate

If you line up first-place rate, you can see if you can improve average placement by increasing the number 1. If average placement is good but first-place rate is low, it may be a stable type that often finishes second. In rules where the top value is high in terms of ranking points, it is worth checking to see if there is room to increase first-place rate.

17-2. average placement＋last-place rate

last-place rate shows a sense of stability that suppresses the lowest ranking. Even if average placement is the same, if last-place rate is high, there may be many top and last places, and if last-place rate is low, there may be a group of 2nd and 3rd place. This is an especially important indicator for rules with heavy Lass penalties.

17-3. average placement + Number of games

Be sure to include the number of games played in average placement. Just by looking at the number 2.20, it is difficult to tell whether it is a 10-game winning streak or a 1,000-game winning streak. Game count is the most important supplementary information to distinguish between short-term upside and long-term trends.

17-4. average placement+Competition level

If you know how many opponents you hit, you can evaluate the difficulty of average placement. If you record the rank and table for a rank match, the range of participants for a set mahjong, and the league you belong to for a competitive league, you can make appropriate comparisons even after the environment changes.

18. Things to keep in mind to improve average placement

18-1. Reduce the lath

For four-player mahjong, 4th place will greatly worsen average placement. Don't just aim to come back to the top when your game has dropped significantly, but also check the realistic conditions for moving up to 3rd place. If you can change 4th place to 3rd place 5 times in 100 games, average placement will improve by 0.05.

However, this does not mean that you should only think about avoiding the last shot and attack passively from the beginning. It is important to aim for additional points in the early stages when the point difference is small, and to judge the value of each finish when the situation becomes solidified in the final stages.

18-2. Reduce excessive pushing

When you receive riichi from your opponent, check how many chantenges you have in your hand, how many points you will score if you win, and whether you have a safe tile. If you press many danger tiles from distant and cheap hands, the loss of shooting tends to be large compared to the reward you gain.

Whether to push or get off cannot be evaluated solely by the result of whether or not the gun was fired. Look back at the status of the round, wait, RBI, and score sticks status at the time of pressing the tile sheet, and create a judgment standard that allows you to choose again under the same conditions.

18-3. Aim well when aiming for the top

It is important to avoid the last-place finish, but if you lean too much on defense, you may increase the number of second places but decrease the number of first places, and average placement may not improve as much as you want. In a competitive game where you can expect to take the lead or score high points, there are times when you need to push to take the lead.

Especially in rules where the top ranking points are high, it is more worthwhile to take a certain amount of risk and aim for a reversal than to finish safely with a close second place. Please take into account the ranking points you are using when making your decision.

18-4. Check the Orus ranking conditions

In Aurus, you check at the beginning of the game how many points you need to score to move up one position. The points required for ron and tsumo may be different, and the order of finishing may change depending on the opponent you hit directly.

If you organize the conditions, including riichi deposits, stacking stick, dealer or non-dealer, you can avoid cheap agari that does not change the ranking or mistakes that cause you to break the necessary moves. In order to accurately judge complex conditions every time, it is effective to practice recalculating after the actual game.

18-5. Change the push and pull depending on the score sticks situation

Even if the tiles are the same, the value of pressing a dangerous tile is different between the top and last tiles. If you're on top, it's worth avoiding big shots to advance the game, and if you're on last, your order won't change even if you land safely, so you may need to take some risks.

Let's think about not only the losses and points conceded by each station, but also how much the results will change the final ranking. If you can push and pull depending on the situation, it will be easier to improve average placement even if the firing rate and firing rate are the same.

18-6. Be more conscious of the final ranking than the one in front of you

In mahjong, it is not always the best thing to do what is wrong. There are situations where playing dealer with a cheap Agari is disadvantageous, and there are situations where there is great value in finishing the station at the top.

If you get into the habit of thinking not only about ``how many points will you gain with this win,'' but also ``what place will you finish after winning'' and ``what ranking will you receive in the next game?'', you will find it easier to make choices that keep your final ranking in mind.

18-7. Don't chase only average placement

Even if you want to improve your average placement, it is counterproductive to change the way you hit each time to match short-term numbers. Even if you make the right decision, you may still end up in last place, and there may be times when your forced push ends up being successful and you come out on top.

Use average placement as an indicator to check the results, and for improvement, look at first-place rate, last-place rate, firing rate, and the contents of the tiles. Increasing the number of decisions that can be repeated in the same situation, rather than going up or down, will lead to long-term improvements in average placement.

20. [Final conclusion] mahjong strength chart of average placement

four-player mahjong

three-player mahjong

21. Summary｜four-player mahjong is 2.50, sanma is 2.00, and looks at "number of games" and "strength of opponent"

The standard for four-player mahjong is 2.50, and the standard for sanma is 2.00. four-player mahjong The first half of 2.4 is clearly good, if you can maintain 2.3 for a long time it's pretty strong, and 2.2 is extremely good. However, even in the M League, while players who are doing well record numbers of 2.1 to 2.2, in matches between top professionals, there are also players who record numbers of 2.5 to 2.6 or higher.

Therefore, it is not possible to draw a simple line such as "average placement2.30=Pro". sanma also has no unified professional standards. Be sure to evaluate average placement based on the number of games, rules, table level, and opponent.

Compared to 2.20 for 10 games, 2.35 maintained over 1,000 and 2,000 games is a number that makes it easier to evaluate reproducibility in terms of skill. first-place rate, last-place rate, top-two rate, together with the ranking distribution, aim for strength that can be maintained over a long period of time rather than just temporary good results.