I know it can be frustrating.
The system technically does not bar a player from ranking up. It only bars a good player from ranking up quickly (You are bared from ranking up in less than hundreds of games.)
One must be aware that your ranking back up quickly because your mmr is high. After testing promos on alternate accounts, I found statistics which suggest that the best out of 10 only affect your rank not reset your mmr (except of course the games you just played which only slightly affect your mmr depending on how many games played in the past). This is why high mmr silver who win 5 out of 5 often get placed in bronze but gain more lp.
The sad part is it often is rough re-climbing without a duo as your high mmr for your division means you have to carry nearly 100% of the time due to lower win rate pairings until/unless your mmr drops to their level. Again, the easiest way to rank up is to use forced low mmr friends and accounts to boost your win rates. Below the higher divisions Ex. Gold-bronze you have allot of smurfing so truly skill doesn't really matter in these divisions. It is all about mmr bouncing or just being above average which means you will generally have solo que about a 51-55% win rate to rank out alone playing allot of games or duo with another exceptional partner(preferably with lowered mmr accounts). Playing above gold however, skill and cs become the main determinant of your rank as you don't generally get paired with drastic feeders. I Have tested this on alts just to bring awareness to the community. You are free to test this for your self the stats will likely remain the same. (You don't need to be a pro player to do this kind of testing. Anyone can given player win rates per division and the win-rate/matchmaking transparency. It is just math and statistics). Right now, I am using this account to solo que only without duo to see (not my win rate or skill level) but the rate of feeders/low win rate players I am given as a percentage and draw a correlation with my win rate percentage.
Stat Edit:
This season account data for the past 15 ranked games by Christmas eve. 2 Thrown out due to flex one thrown out due to remake (Smite-less Jg and AFK was the cause.)
Silver 2 mmr placed in bronze 2 at 26 lp per win 14 per loss out of 12 ranked solo que games
My average performance
0.666667 decent or carry game-play (good performance)
Number of Team mates out of 5 with a good average performance
2.416667
Number of Team mates out of 5 with a bad average performance
2.583333
Number of enemies out of 5 with a good average performance
2.916667
Number of enemies of 5 with a bad average performance
2.083333
1 game out of 13 Jg no smite Remake (threw out two ranked flex)
This means I have slightly over 50% chance to be paired with bad team mates making my team somewhat fair. The enemies however had slightly better team mates which I will attribute to an unlucky encounter with smurfs or that riot made them stronger to account for my mmr. To prevent bias, I have not stated with win or losses of the games so the team pairing quality alone can be viewed for this solo que account. needless to say I lost most of these games as the ratio of good team mates:good enemies was 2.4:2.9 which suggests a large correlation to KDA performance and winning which is good. Because this is bronze I did not factor cs into performance yet KDA alone was enough to predict the win loss ratio which is below 50% for this season. (I imagine this is different in higher divisions where cs matters more) Yet in bronze at least, you can actually complain about KDA as a factor in determining the win.
Stay informed!
